Photoisomerization of bicyclopropylidene and 1,2-dimethylenecyclobutane in rare-gas matrices: Towards the IR-spectroscopic identification of tetramethyleneethane (2,3-dimethylenebutane-1,4-diyl)

Article abstract of DOI:10.1002/(SICI)1099-0690(199906)1999:6<1291::AID-EJOC1291>3.0.CO;2-M

Bicyclopropylidene (3) and 1,2-dimethylenecyclobutane (7) have been irradiated in rare-gas matrices. If 1,2-dimethylenecyclobutane (7) is exposed to the light of a KrF excimer laser (λ = 248 nm), an isomeric species is produced, showing an absorption at 793.1 cm^-1 (argon matrix) or 791.2 cm^{- 1} (xenon matrix) in the IR spectrum. The back reaction can be induced with light of k λ = 254 nm. This photochemical interconversion, together with the comparison between the experimental and calculated band positions, supports the assignment of the IR absorption near 790 cm^-1 to tetramethyleneethane (5).

Full text of DOI:10.1002/(SICI)1099-0690(199906)1999:6<1291::AID-EJOC1291>3.0.CO;2-M

FULL PAPER
Photoisomerization of Bicyclopropylidene and 1,2-Dimethylenecyclobutane in
Rare-Gas Matrices: Towards the IR-Spectroscopic Identification of
Tetramethyleneethane (2,3-Dimethylenebutane-1,4-diyl)
Günther Maier*[a] and Stefan Senger[a]
Keywords: Matrix isolation / IR spectroscopy / Photochemistry / Diradicals / Isomerizations
Bicyclopropylidene (3) and 1,2-dimethylenecyclobutane (7)
have been irradiated in rare-gas matrices. If 1,2-
dimethylenecyclobutane (7) is exposed to the light of a KrF
excimer laser (λ = 248 nm), an isomeric species is produced,
showing an absorption at 793.1 cm–1 (argon matrix) or 791.2
cm–1 (xenon matrix) in the IR spectrum. The back reaction
can be induced with light of λ = 254 nm. This photochemical
interconversion, together with the comparison between the
experimental and calculated band positions, supports the
assignment of the IR absorption near 790 cm–1 to
tetramethyleneethane (5).
Recently, we succeeded in generating 4-methylene-2-pen- trix experiments bicyclopropylidene (3) and 1,2-dimethyl-
tene-1,5-diyl (2) in a xenon matrix and were able to charac- enecyclobutane (7) were selected.
terize this species IR-spectroscopically.[1] This success was
made possible by the application of a new method for gen-
erating highly reactive species in rare-gas matrices that we
have developed: 2 was formed in situ by irradiation of 1-
Matrix Experiments
Bicyclopropylidene (3)
methylene-2-vinylcyclopropane (1) in a xenon matrix ad-
ditionally doped with bromine atoms. Our method evolved
on the basis of experiences we gained in connection with
the IR-spectroscopic detection of trimethylenemethane.[2]
Whereas 1 is photostable if irradiated at λ ϭ 313 nm in
a xenon matrix, photolysis of 1 at the same wavelength in
a bromine-doped xenon matrix leads to the formation of
the diradical 2.
In a xenon matrix at 10 K, bicyclopropylidene (3) was
exposed to light of wavelength λ ϭ 313 nm. Even after 20
h, no change in the matrix IR spectrum could be detected.
Thus, 3 proved to be photostable under these conditions.
However, the result was completely different when the
xenon matrix was additionally doped with bromine (ratio
3/Br2/Xe ϭ 1.5:1.5:1000). After just 1 h, the intensities of
the IR absorptions of bicyclopropylidene (3) had decreased
by 5.5%, and after 15 h, the degree of conversion was ap-
proximately 47%. 1,2-Dimethylenecyclobutane (7) was
formed almost exclusively, with just a small amount of
methylenespiropentane (6) additionally being detected
[Scheme 1, path (a)].
It is most likely that the initial product in the conversion
of bicyclopropylidene (3) to 1,2-dimethylenecyclobutane (7)
is the singlet diradical S-4. 2-Cyclopropylidenepropane-1,3-
diyl (4) is a member of the series of trimethylenemethane
diradicals and therefore should have a triplet ground state,
as is the case for trimethylenemethane[6] itself and many of
its derivatives.[7] In this context, we assumed that in the
course of the irradiation of 3, T-4 could also be formed
from S-4 by ISC.
4-Methylene-2-pentene-1,5-diyl (2) is also designated as a
1,2Ј-bis(allyl) diradical since it can formally be obtained by
the combination of two allyl radicals under hydrogen ab-
straction. Accordingly, tetramethyleneethane (5) is a 2,2Ј-
bis(allyl) diradical. Furthermore, 5 represents one of the
[3,4]
´
simplest non-Kekule hydrocarbons.
Even though more than 25 years have elapsed since
Dowd[5] succeeded in characterizing tetramethyleneethane
(5) by means of ESR spectroscopy, as yet this species has
not been detected IR-spectroscopically. Encouraged by the
successful identification of the 1,2Ј-bis(allyl) diradical 2, we
decided to direct our attention towards the 2,2Ј-bis(allyl)
diradical 5 in the hope that application of our new method
would prove fruitful. As promising precursors for the ma-
In the geometry calculated for the triplet state of 4 by the
UB3LYP/6-31G* method, all carbon atoms lie in the same
plane such that the molecule possesses C2v symmetry. Ex-
actly as for trimethylenemethane,[2] by far the most intense
absorption of T-4 is that due to the ωCH2 vibration of the
olefinic methylene groups. In the IR spectrum of T-4 calcu-
lated by the UB3LYP/6-31G* method, this vibration is lo-
cated at 718.2 cmϪ1. Correction of the calculated value by
a factor of 1.011, determined in the case of trimethylene-
[a]
Institut für Organische Chemie der Justus-Liebig-Universität,
Heinrich-Buff-Ring 58, D-35392 Giessen, Germany
Fax: (internat.) ϩ 49(0)641/99-34309
Eur. J. Org. Chem. 1999, 1291Ϫ1294
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1434Ϫ193X/99/0606Ϫ1291 $ 17.50ϩ.50/0
1291

G. Maier, S. Senger
FULL PAPER
the UV spectrum of a solution of 7 in isooctane is reported
to be at λ ϭ 246 nm.[9]
As the product of irradiation at λ ϭ 254 nm, 3-methyl-
enecyclopentene (11) was formed almost exclusively. In ad-
dition, only weak absorptions due to ethylene (9) and buta-
triene (10) could be observed in the IR spectrum. Allene,
which is found as a photoproduct of 7 upon irradiation at
λ ϭ 254 nm in the gas phase,[10] could not be observed in
the matrix. Likewise, no absorptions indicating the interme-
diacy of tetramethyleneethane (5) could be detected.
The formation of 3-methylenecyclopentene (11), ethylene
(9), as well as butatriene (10) can be rationalized in terms
of the formation of diradical 8 upon cleavage of the C2ϪC3
bond in 1,2-dimethylenecyclobutane (7) in the initial reac-
tion step. Through the formation of a five-membered ring
and a subsequent hydrogen shift, the diradical 8 can react
further to give 11, whereas a second CϪC bond cleavage
leads to 9 and 10.
1,2-Dimethylenecyclobutane (7) was also exposed to the
light of a KrF excimer laser (λ ϭ 248 nm) in both argon
and xenon matrices [Scheme 1, path (c)]. Again, 3-methyl-
enecyclopentene (11) was found as the main product of the
photolysis. In contrast to the irradiation at λ ϭ 254 nm, no
absorptions due to ethylene (9) or butatriene (10) could be
detected in the IR spectrum after the laser photolysis of 7.
Besides the absorptions due to 11, several weak bands
were seen in the difference spectrum (xenon matrix: 791.2,
1191.2, 1670.6, 1761.7 cmϪ1, cf. Figure 1). Of special inter-
est is the absorption found at 791.2 cmϪ1 in a xenon matrix
and at 793.1 cmϪ1 in an argon matrix. This absorption
would appear to be a very promising candidate for assign-
ment to the elusive tetramethyleneethane (5) as it is posi-
tioned in the same region of the IR spectrum as the most
intense vibration of the structurally related allyl radical. In
an argon matrix, the most intense absorption of the allyl
Scheme 1. Survey of the reaction pathways of bicyclopropylidene
(3) and 1,2-dimethylenecyclobutane (7) on irradiation in rare-gas
matrices; irradiation conditions: (a) at λ ϭ 313 nm in a bromine-
doped xenon matrix, (b) at λ ϭ 254 nm in an argon or xenon
matrix, (c) with a KrF excimer laser (λ ϭ 248 nm) in an argon or
xenon matrix
radical is seen at 802 cmϪ1 [11]
.
methane, results in a band near 726 cmϪ1. However, in our
matrix experiments no product absorptions were detected
in this spectral region. Likewise, there was no experimental
indication that tetramethyleneethane (5) had been formed.
The band positions of methylenespiropentane (6)[8] in a
xenon matrix have been observed by flash pyrolysis of a
bicyclopropylidene (3)/xenon gas mixture and condensation
of the products on a spectroscopic window at 10 K. In the
course of a subsequent irradiation at λ ϭ 254 nm, 1,2-di-
methylenecyclobutane (7) and 3-methylenecyclopentene
(11) were formed as photoproducts.
Figure 1. IR difference spectrum of the laser irradiation (λ ϭ 248
nm) of 1,2-dimethylenecyclobutane (7) in a xenon matrix [Scheme
1, path (c)]; the bands marked * could not be assigned; the bands
with positive values were enhanced upon irradiation
1,2-Dimethylenecyclobutane (7)
Unlike bicyclopropylidene (3), 1,2-dimethylenecyclo-
butane (7) proved to be photostable on irradiation at λ ϭ
Following the laser photolysis at λ ϭ 248 nm, the argon
313 nm in a bromine-doped xenon matrix. If light of the and xenon matrices were exposed to the light of a low-
wavelength of λ ϭ 254 nm was used instead, a photoreac- pressure mercury lamp equipped with a vycor filter (λ ϭ
tion was observed in the undoped xenon matrix as well as 254 nm). As a result, the intensity of the IR band at 791.2
in an argon matrix, because 7 is able to absorb light of this cmϪ1 (xenon matrix) or 793.1 cmϪ1 (argon matrix) de-
wavelength directly [Scheme 1, path (b)]. The maximum in creased rapidly. As the sole product, 1,2-dimethylenecyclo-
1292
Eur. J. Org. Chem. 1999, 1291Ϫ1294

Photoisomerization of Bicyclopropylidene and 1,2-Dimethylenecyclobutane in Rare-Gas Matrices
FULL PAPER
butane (7) could be detected by means of the difference angles between the allylic subunits were calculated as 44.8°
spectrum (see Figure 2). If the matrices were then further (CAS) and 44.9° (UB3LYP) in S-5. For T-5, we calculated
irradiated with the KrF excimer laser (λ ϭ 248 nm), the slightly larger dihedral angles of 50.3° (CAS) and 47.7°
intensity of the absorption at 791.2 cmϪ1 (xenon matrix) or (UB3LYP).
793.1 cmϪ1 (argon matrix) increased once more.
Table 1. Calculated IR absorptions and relative intensities of singlet
and triplet tetramethyleneethane S-5 and T-5
UB3LYP/6-31G*
CAS(6,6)/6-31G*
S-5
T-5
S-5
T-5
ν˜ [cmϪ1
]
int.
ν˜ [cmϪ1
]
int.
ν˜ [cmϪ1
]
ν˜ [cmϪ1
]
symmetry
271.4 < 0.01 264.8 < 0.01
281.8
314.5
271.2
298.1
B2
B1
B3
B2
B1
B3
B1
B2
B3
B1
B2
B1
B2
B3
B2
B3
B1
B2
B1
B3
B1
B2
B3
B1
B3
B2
304.4
462.7
552.4
553.8
0.02 293.7
0.02 461.4
0.01 555.4
0.04 553.9
0.02
0.02
0.02
0.04
492.4
490.1
572.5
576.6
577.5
579.4
558.2 < 0.01 561.8 < 0.01
586.7
589.4
562.8
684.2
760.9
795.8
803.4
0.07 567.5
0.03 672.5
0.01 764.8
1.00 797.7
0.10 806.2
0.06
0.01
0.01
1.00
0.13
584.7
588.6
694.1
681.9
689.9
693.3
722.9
722.2
Figure 2. IR difference spectrum of the photochemically induced
back reaction at λ ϭ 254 nm (40 min) after previous irradiation of
1,2-dimethylenecyclobutane (7) with a KrF excimer laser (λ ϭ 248
nm); the bands with positive values diminished, while those with
negative values were enhanced upon irradiation
728.1
727.1
993.8 < 0.01 992.4 < 0.01
995.8 < 0.01 994.4 < 0.01
1045.9
1055.7
1094.6
1262.9
1371.6
1372.0
1620.8
1614.8
1658.4
3336.4
3338.3
3345.9
3433.8
3437.2
3437.7
1040.3
1048.8
1096.8
1237.4
1371.7
1346.9
1616.3
1611.4
1658.1
3335.1
3337.2
3344.3
3432.2
3435.4
3435.2
1045.4
1285.2
1300.5
0.01 1047.5
0.03 1275.3
0.02 1301.3
0.01
0.04
0.02
1362.6 < 0.01 1353.9 < 0.01
Thus, the species responsible for the absorption near 790
cmϪ1 can be generated from 1,2-dimethylenecyclobutane (7)
by laser photolysis at λ ϭ 248 nm and can subsequently be
transformed back into 7 by exposing the matrix to light of
wavelength λ ϭ 254 nm. This behavior strongly supports
the assignment of the 790 cmϪ1 absorption to tetramethyl-
eneethane (5).
1509.9
1509.9
1544.2
3170.0
3171.5
3176.7
3265.7
3267.6
3268.6
0.09 1506.8
0.01 1506.5
0.08 1544.6
0.02 3169.0
0.07 3170.3
0.16 3175.8
0.01 3264.9
0.26 3266.8
0.03 3267.6
0.09
0.01
0.08
0.02
0.07
0.17
0.02
0.27
0.03
It seems rather surprising that tetramethyleneethane (5)
could only be detected in the IR spectrum when 1,2-dimeth-
ylenecyclobutane (7) was irradiated with the excimer laser
(λ ϭ 248 nm), whereas photolysis with the low-pressure
mercury lamp (λ ϭ 254 nm) did not lead to the formation
of this species. Apparently, the high photon density of the
laser is essential to bring about this photoreaction.
As mentioned above, methylenespiropentane (6) yields
1,2-dimethylenecyclobutane (7) upon irradiation at λ ϭ 254
nm in a xenon matrix. Therefore, 6 does not represent a
valid candidate for an independent approach to 5 since its
irradiation will invariably give rise to 7 as an intermediate.
In the calculated IR spectra of S-5 and T-5, the wagging
vibration of the methylene groups (ωCH2) gives by far the
most intense band. The calculated IR spectral bands are
listed in Table 1. The results obtained with CAS(6,6) and
UB3LYP lead to the prediction that in the IR spectra of
both S-5 and T-5, the most intense absorption lies in close
proximity to the corresponding vibration of the allyl rad-
ical. Since the calculated band positions for the singlet and
triplet states of tetramethyleneethane (5) differ by less than
2 cmϪ1, it is not possible to determine the multiplicity of
this species through comparison of the calculated and ex-
perimental frequencies.
If the band position of the ωCH2 vibration of the allyl
radical is computed using the UB3LYP/6-31G* approach,
IR Spectrum
The theoretical IR spectrum of tetramethyleneethane (5) the result (797.0 cmϪ1) is in extremely satisfying accordance
has not previously been reported in the literature. Conse- with the experimental value (802 cmϪϪ1).[11] This corre-
quently, we performed ab initio calculations with the Gaus- sponds very well with the findings we made in studying the
sian 94 package of programs[12] using the CAS(6,6) and 1,2Ј-bis(allyl) diradical 2. In the case of 2, the deviation of
UB3LYP methods in combination with the basis set 6-31G* the theoretical value from the experimentally determined
. Since the RB3LYP wavefunction for S-5 was not stable, band position of the ωCH2 vibration was only about 10
we had to perform the calculations for the singlet state of cmϪ1. In order to assess the suitability of the UB3LYP
tetramethyleneethane (5) with UB3LYP and the keyword method for open-shell singlet molecules such as S-5, we
“guessϭmix”. With both theoretical approaches [CAS(6,6) compared the calculated and experimental IR spectra of
and UB3LYP], singlet-tetramethyleneethane (S-5) was pre- singlet-1,2,4,5-tetramethylenebenzene
[2,3,5,6-tetrakis-
dicted to be the ground state (∆EST ϭ ET Ϫ ES ϭ ϩ1.3 and (methylene)-1,4-cyclohexanediyl]. Berson et al.[13] reported
ϩ0.9 kcal molϪ1, respectively) and the localized minima ob- the most prominent features in the matrix IR spectrum (ar-
tained for S-5 and T-5 were of D2 symmetry. The dihedral gon matrix) of singlet-1,2,4,5-tetramethylenebenzene to ap-
Eur. J. Org. Chem. 1999, 1291Ϫ1294
1293

G. Maier, S. Senger
FULL PAPER
pear at 807.6 and 849.3 cmϪ1. Roth, Maier et al.[14] as-
signed two absorptions at 810 and 854 cmϪ1 (argon matrix)
to 1,2,4,5-tetramethylenebenzene. At the UB3LYP/6-31G*
level of theory, the two most intense vibrations of singlet-
1,2,4,5-tetramethylenebenzene (D2h symmetry) are located
at 804.2 and 859.6 cmϪ1, in excellent agreement with the
experimental data. With UB3LYP/6-31G*, the singlet state
of 1,2,4,5-tetramethylenebenzene is calculated to be 4.1 kcal
molϪ1 more stable than the triplet, a result also in accord-
ance with experiment.[13] Therefore, application of the UB-
3LYP method may be expected to offer a very reliable pre-
diction of the position of the ωCH2 vibration of not only
T-5 but also S-5.
the matrix window at 10 K as gas mixtures with xenon or argon
(ratio 3:1000). The most intense absorption of 3 in a xenon matrix
was located at 952.5 cmϪ1. In the case of 7, the main band was
seen at 875.6 cmϪ1. The bromine-doped matrices were prepared by
co-condensation of the 3/xenon gas mixture with a bromine/xenon
gas mixture (ratio 3:1000) in a 1:1 ratio.
Calculations: The calculations were performed with the Gaussian
94 package of programs.[12] The 6-31G* basis set was used through-
out. For the DFT calculations of S-5, UB3LYP was used in combi-
nation with the keyword “guessϭmix”. For these calculations, the
stability of the wavefunction was tested with the “stable” keyword.
The energies E and zero-point energies ZPE (au) calculated for 5
and 1,2,4,5-tetramethylenebenzene were as follows. CAS(6,6)/6-
31G*: S-5 (D2 symmetry, dihedral angle: 44.8°, E ϭ Ϫ231.8135643,
Our calculations lead to the conclusion that the most in- ZPE ϭ 0.119171), T-5 (D2 symmetry, dihedral angle: 50.3°, E ϭ
Ϫ231.81139, ZPE ϭ 0.119011). UB3LYP/6-31G*: S-5 (D2 sym-
metry, dihedral angle: 44.9°, E ϭ Ϫ233.3258507, ZPE ϭ 0.114328),
T-5 (D2 symmetry, dihedral angle: 47.7°, E ϭ Ϫ233.324513, ZPE ϭ
0.114349), singlet-1,2,4,5-tetramethylenebenzene (D2h symmetry,
E ϭ Ϫ386.9648356, ZPE ϭ 0.161541), triplet-1,2,4,5-tetramethyl-
enebenzene (D2h symmetry, E ϭ Ϫ386.9580914, ZPE ϭ 0.161294).
tense vibration of tetramethyleneethane (5) should be lo-
cated near 800 cmϪ1. This is a further strong argument in
favor of our assignment of the absorption found in the ex-
perimental IR spectrum near 790 cmϪ1 to species 5.
Results and Conclusions
Acknowledgments
On irradiation of matrix-isolated 1,2-dimethylenecyclo-
butane (7) with a KrF excimer laser (λ ϭ 248 nm), an ab-
sorption near 790 cmϪ1 (argon matrix: 793.1 cmϪ1, xenon
matrix: 791.2 cmϪ1) has been observed in the IR spectrum.
Subsequent exposure of the matrix to the light of a low-
pressure mercury lamp (λ ϭ 254 nm) led to a decrease in
the intensity of this IR band. By means of the IR difference
spectrum, 7 was identified as the sole product of the photo-
chemically induced back reaction. This experimental result
leads to the conclusion that the species responsible for the
absorption near 790 cmϪ1 has the elemental composition
C6H8.
The band position calculated (UB3LYP/6-31G*) for the
most intense vibration of tetramethyleneethane (5) is in in
very good agreement with the experimentally determined
band position of the C6H8 species. Although we are well
aware that it is not possible to unequivocally characterize a
chemical species on the basis of only one IR-active vi-
bration, we assign the IR absorption at 793.1 cmϪ1 in an
argon matrix or at 791.2 cmϪ1 in a xenon matrix to the
ωCH2 vibration of tetramethyleneethane (5).
We wish to thank the Deutsche Forschungsgemeinschaft and the
Volkswagen-Stiftung for their support.
[1]
G. Maier, S. Senger, J. Am. Chem. Soc. 1997, 119, 5857Ϫ5861.
[2]
[2a] G. Maier, H. P. Reisenauer, K. Lanz, R. Troß, D. Jürgen, B.
A. Hess, Jr., L. J. Schaad, Angew. Chem. 1993, 105, 119Ϫ121;
[2b]
Angew. Chem. Int. Ed. Engl. 1993, 32, 74Ϫ76. Ϫ
G. Maier,
D. Jürgen, R. Troß, H. P. Reisenauer, B. A. Hess, Jr., L. J.
Schaad, Chem. Phys. 1994, 189, 383Ϫ399.
[3]
[4]
J. A. Berson, Acc. Chem. Res. 1997, 30, 238Ϫ244.
D. A. Hrovat, W. T. Borden, J. Mol. Struct. (Theochem.) 1997,
398Ϫ399, 211Ϫ220.
[5]
[6]
P. Dowd, J. Am. Chem. Soc. 1970, 92, 1066Ϫ1068.
R. J. Baseman, D. W. Pratt, M. Chow, P. Dowd, J. Am. Chem.
Soc. 1976, 98, 5726Ϫ5727.
[7]
J. A. Berson in Diradicals (Ed.: W. T. Borden), John Wiley &
Sons, New York, 1982, pp. 151Ϫ194.
[8]
[9]
W. R. Dolbier, Jr., Tetrahedron Lett. 1968, 393Ϫ396.
J. J. Gajewski, C. N. Shih, J. Am. Chem. Soc. 1972, 94,
1675Ϫ1682.
[10]
[11]
[12]
P. A. Kelso, A. Yeshurun, C. N. Shih, J. J. Gajewski, J. Am.
Chem. Soc. 1975, 97, 1513Ϫ1519.
G. Maier, H. P. Reisenauer, B. Rohde, K. Dehnicke, J. Am.
Chem. Soc. 1983, 116, 732Ϫ740.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Peters-
son, J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V.
G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B.
Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y.
Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R.
Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees,
J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, J. A.
Pople, Gaussian 94, Revision B.1, Gaussian Inc., Pittsburgh,
Pennsylvania, 1995.
Experimental Section
General: The cryostat used for matrix isolation was a Displex CSA
202 closed-cycle refrigeration system from Air Products. The tem-
perature measurement and control were performed with a digital
temperature indicator/controller 3700-APD-E from Air Products
[gold (0.07% iron)/chromel thermo element]. The matrix window
consisted of CsI and spectra were obtained with a Bruker IFS 85
FT-IR spectrometer. The light sources used were a low-pressure
spiral mercury lamp with a vycor filter and an LPX 105 MC ex-
cimer laser from Lambda Physics. Ϫ Bicyclopropylidene (3) was
prepared according to the procedure of de Meijere et al.[15] 1,2-
Dimethylenecyclobutane (7) was prepared from 1,2-bis(hydroxy-
methyl)cyclobutane.[9] Ϫ Compounds 3 and 7 were condensed onto
[13]
[14]
J. H. Reynolds, J. A. Berson, K. K. Kumashiro, J. C. Duchamp,
K. W. Zilm, J. C. Scaiano, A. B. Berinstain, A. Rubello, P. Vo-
gel, J. Am. Chem. Soc. 1993, 115, 8073Ϫ8090.
W. R. Roth, R. Langer, M. Bartmann, B. Stevermann, G. Ma-
ier, H. P. Reisenauer, R. Sustmann, W. Müller, Angew. Chem.
1987, 99, 271Ϫ272; Angew. Chem. Int. Ed. Engl. 1987, 26,
256Ϫ258.
[15]
A. de Meijere, S. I. Kozhushkov, T. Spaeth, N. S. Zefirov, J.
Org. Chem. 1993, 58, 502Ϫ505.
Received November 30, 1998
[O98540]
1294
Eur. J. Org. Chem. 1999, 1291Ϫ1294