Addition of dichloro- and dibromocarbene to 1,2-diphenylcyclobutene

Article abstract of DOI:10.1246/cl.2000.246

1,2-Diphenylcyclobutene (7) was reacted with dibromo- and dichlorocarbene, both generated via three different methods. 1,3-Diphenyl-2-halocyclopenta-1,3-dienes 12 were isolated which result from addition of the dihalocarbenes to the cyclobutene double bond of 7. A cationic cyclopropyl-allyl rearrangement (CCA) in gem-dihalobicyclopentanes 8 leads to 2,3-dihalocyclopentenes 9, which under the reaction conditions are dehydrohalogenated to 12. A second carbene addition and rearrangement afford aromatic compounds 11 and 16.

Full text of DOI:10.1246/cl.2000.246

246
Chemistry Letters 2000
Addition of Dichloro- and Dibromocarbene to 1,2-Diphenylcyclobutene
Robert A. Wagner, Jürgen Weber, and Udo H. Brinker*
Institut für Organische Chemie, Unversität Wien, Währinger Str. 38, A-1090 Wien, Austria
(Received November 29, 1999; CL-991013)
differential orbital energies (LUMOcarbene–HOMOalkene).4
Consequently, based on determination of differential orbital
energies,15 one might predict a similar reactivity for the addi-
tion of dihalocarbenes to 1,2-diphenylcyclobutene (7).
1,2-Diphenylcyclobutene (7) was reacted with dibromo-
and dichlorocarbene, both generated via three different meth-
ods. 1,3-Diphenyl-2-halocyclopenta-1,3-dienes 12 were isolat-
ed which result from addition of the dihalocarbenes to the
cyclobutene double bond of 7. A cationic cyclopropyl-allyl
rearrangement (CCA) in gem-dihalobicyclopentanes 8 leads to
2,3-dihalocyclopentenes 9, which under the reaction conditions
are dehydrohalogenated to 12. A second carbene addition and
rearrangement afford aromatic compounds 11 and 16.
The most common and thoroughly investigated reaction of
carbenes is the addition to carbon-carbon double bonds.
Although vast literature concerning dihalocarbene reactions
with open chain and cyclic alkenes larger than four-membered
rings1-3 exists, only a few studies with small-ring alkenes have
been reported.4-6
Only very few dihalocarbene additions to cyclobutenes are
known.5,6 To provide further evidence for the proposed mecha-
nism (Scheme 1), 1,2-diphenylcyclobutene (7)16 was treated
with dibromo- and dichlorocarbene (Scheme 2). The dihalocar-
benes were generated by three different ways, using the meth-
ods of Doering and Hoffmann (KOtBu/CHX3),17 Seyferth
(PhHgCX3/refluxing benzene)18 and Xu (ultrasound).19,20
The gem-dihalovinylcyclopropanes 14 (Scheme 2) which,
among other compounds, were expected to be formed in the
reactions, were synthesized independently. To this end, Horner-
Wadsworth-Emmons reactions were performed on (1-phenylcy-
clopropyl)phenyl ketone21 and the corresponding diethyl-1,1-
dihalomethylphosphonate22 at room temperature which afford-
ed 14 (a and b) in 13% yield each.
Since the pioneering work of Skell and Woodworth7,8 and
Doering,9 it is generally accepted that the singlet carbene addi-
tion occurs by a one-step process in which two new σ-bonds
are formed simultaneously. Although this type of addition pro-
ceeds in a concerted fashion, it cannot be synchronous, due to
orbital symmetry considerations.1,10
In 1956, Skell and Garner11 suggested that the addition of
dibromocarbene to several alkenes would lead to a transition
state with complete charge separation, but the intermediacy of a
complex “as a partially formed cyclopropane that has some car-
bonium ion character developed on one of the carbons of the
double bond” was never proven. Later on, Yang and
Marolewski12 proposed a stepwise addition of photochemically
generated monoiodo- and monochlorocarbenes to 1,2-dimethyl-
cyclobutene in order to explain the formation of the 2-(1-
methylcyclopropyl)-1-halopropenes found. Jones et al.13 sug-
gested for this reaction the involvement of a biradical, which
would result from addition of a triplet carbene. Due to the small
energy gap between the singlet ground state and the triplet state
of monohalo- and some dihalocarbenes, an equilibrium was
postulated. In order to explain the formation of the products,
however, the rate of reaction of the triplet carbene must be
much larger than that of the singlet.
Recently, for the reactions of dihalocarbenes with 1,2-di-
phenylcyclopropenes 1, a two-step mechanism involving dipo-
lar or polarized activated complexes, such as 2, has been sug-
gested (Scheme 1) to explain the formation of the 1,3-butadi-
enes 3.4
Moreover, 1 can react with dihalocarbenes to afford the
putative bicyclo[1.1.0]butane 4. After cleavage of the central
bond in 4 and loss of one halide, the homoaromatic
cyclobutenyl cation 5 is generated.14 Recombination of 5 with
the halogen anion gives cyclobutene 6 as the major product.
The high reactivity of strained cyclopropene 1 towards
dihalocarbenes is clearly reflected by the narrow gaps of the
In the dichloro- and dibromocarbene additions to 1,2-
diphenylcyclobutene (7), the intermediate adduct 8 could never
be isolated nor spectroscopically observed without doubt.
Instead, 1,3-diphenyl-2,3-dihalocyclopentene (9), deriving from
a CCA rearrangement 8 → 9,10,14 is believed to be formed.
Under all three conditions applied, however, 9 is rapidly dehy-
drohalogenated to afford 1,3-diphenyl-2-halocyclopenta-1,3-
diene (12). In addition, the hitherto unknown aromatic com-
pounds 11 and 16 were produced. Therefore, under these condi-
tions, cyclopentadiene 12 underwent a second carbene addition.
This reaction is favored at the sterically less hindered and more
Copyright © 2000 The Chemical Society of Japan

Chemistry Letters 2000
247
electron-rich double bond, yielding 1,3-diphenyl-2,6,6-trihalo-
bicyclo[3.1.0]hex-2-ene (10). It occurs also, in lower yield, at
the other double bond of 12 to give 2,5-diphenyl-1,6,6-trihalo-
bicyclo[3.1.0]hex-2-ene (15). After ring opening and concomi-
tant dehydrohalogenation, 10 is converted to 11. The aromatiza-
tion of 15 → 16 proceeds in the same manner.
gy are released, almost all strain energy (ca. 51 kcal/mol) is set
free in the corresponding rearrangement of bicyclopentane to
cyclopentene. Moreover, if the carbene addition to 7 would pro-
ceed via 13, i.e. in a stepwise fashion, as has been suggested for
1,4 to give gem-dihalovinylcyclopropanes 14, almost no ring
strain is lost. In contrast, there is considerable driving force in
the corresponding reaction 2 → 3, due to the formation of
strain-free butadienes 3. These energy considerations clearly
show the exceptionally high reactivity of cyclopropenes and the
importance of ring strain in the reaction pathways of carbene
additions.
In a typical experiment, under Doering-Hoffmann condi-
tions no reaction occured, and starting material 7 was recovered
completely. In contrast, when ultrasound was applied to the
reaction mixture,20 after 35 min about 30% of 11a, 10% of 16a
and small amounts of 1,3-diphenyl-2,6,6-trichlorobicy-
clo[3.1.0]hex-2-ene (10a) could be isolated (Table 1, entry 1).
If, under the same conditions, ultrasonication was extended to
60 min (entry 3), almost no starting material could be detected,
but 45% of 11a along with 26% of 16a were observed. In order
to avoid basic conditions and prevent any subsequent dehydro-
halogenation of 9, 10 and 15, Seyferth´s method was used
(entries 4 - 6). Still, no 1,3-diphenyl-2,3-dihalocyclopentenes
(9) could be detected. Interestingly, 10a could be isolated as
main product (entry 5), whereas with PhHgCBr3 only the aro-
matic compound 11b was formed (entry 6). Bromide is a better
leaving group than chloride and that might be responsible for
this result.
References and Notes
1
W. M. Jones and U. H. Brinker, in “Pericyclic Reactions,” ed
by A. P. Marchand and R. E. Lehr, Academic Press, New York
(1977), p. 110.
2
3
R. A. Moss, Acc. Chem. Res., 22, 15 (1989), and Refs. 29 and
31 cited therein.
E. V. Dehmlow, in “Houben-Weyl: Methoden der Organischen
Chemie,” ed by M. Regitz, Thieme, Stuttgart (1989), Vol. E19b,
p. 1461.
4
5
6
7
8
9
J. Weber and U. H. Brinker, Angew. Chem., Int. Ed. Engl. 36,
1623 (1997), and Refs. 8 and 9 cited therein.
J. Weber, L. Xu, and U. H. Brinker, Tetrahedron Lett., 33, 4537
(1992), and Refs. 2d, 2e and 2f cited therein.
S. B. Lewis and W. T. Borden, Tetrahedron Lett., 35, 1357
(1994).
P. S. Skell and R. C. Woodworth, J. Am. Chem. Soc., 78, 4496
(1956), and correction: J. Am. Chem. Soc., 78, 6427 (1956).
P. S. Skell and R. C. Woodworth, J. Am. Chem. Soc., 81, 3383
(1959).
W. v. E. Doering and P. LaFlamme, J. Am. Chem. Soc., 78,
5447 (1956).
10 R. Hoffmann and R. B. Woodward, Angew. Chem., Int. Ed.
Engl., 8, 781 (1969).
11 P. S. Skell and A. Y. Garner, J. Am. Chem. Soc., 78, 5430
(1956).
12 N. C. Yang and T. A. Marolewski, J. Am. Chem. Soc., 90, 5644
(1968).
13 M. Jones jr., P. P. Gaspar, J. B. Lambert, and V. J. Tortorelli,
Tetrahedron Lett., 1978, 4257, and Refs. 13, 24 and 25 cited
therein.
It is noteworthy, that no gem-dihalovinylcyclopropane 14
resulting from a rearrangement of 13 could be observed with
any of the examined dihalocarbene additions to 1,2-diphenylcy-
clobutene (7). While there is evidence that dihalocarbenes can
undergo a stepwise addition to 1,2-diphenylcyclopropenes 1,4
bearing electron-pushing or electron-pulling substituents at the
para position of one phenyl group, at this point, there is no
support for such a mechanism in the corresponding carbene
reactions with 1,2-diphenylcyclobutene (7). The structural dif-
ferences between 1 and 723 could certainly be one reason. A
stronger argument, however, is based on the enormous differ-
ence in strain energy of parent cyclopropene (55.2 kcal/mol24)
and cyclobutene (28.4 kcal/mol24). In the reaction of dihalocar-
benes with cyclobutene forming bicyclo[2.1.0]pentane, the ring
strain increases by 26 kcal/mol; on the other hand, the calculat-
ed increase for cyclopropene forming bicyclo[1.1.0]butane is
only about 9 kcal/mol. The larger build-up of strain energy in
the case of cyclobutene 7 seems to impede the ease of this reac-
tion. Indeed, when compared with the reaction of dichlorocar-
bene and 1, under otherwise identical conditions, a slower reac-
tion was observed for the corresponding addition to 7.
Furthermore, while in the concomitant rearrangement of bicy-
clobutane to cyclobutene about 36 kcal/mol of total strain ener-
14 C. H. DePuy, Acc. Chem. Res., 1, 33 (1968).
15 J. Weber, Ph. D. Thesis, SUNY-Binghamton, NY, USA, 1995.
16 1,2-Diphenylcyclobutene (7) was synthesized as follows: cyclo-
propanation of desoxybenzoin was performed by use of the sys-
tem NaNH2/DMSO/1-bromo-2-chloroethane. The resulting (1-
phenylcyclopropyl)phenyl ketone was reacted with tosylhy-
drazide in ethanol to yield the corresponding tosylhydrazone,
which was refluxed with sodium methoxide in diglyme at 120
°C for 20 minutes to give compound 7 in an overall yield of
33%.
17 W. v. E. Doering and A. K. Hoffmann, J. Am. Chem. Soc., 76,
6162 (1954).
18 D. Seyferth, Acc. Chem. Res., 5, 65 (1972).
19 L. Xu, W. B. Smith, and U. H. Brinker, J. Am. Chem. Soc., 114,
783 (1992).
20 L. Xu and U. H. Brinker, in “Sonochemical Organic Synthesis,”
ed by J. L. Luche, Plenum, New York (1998), p. 354.
21 M. S. Newman and G. Kaugars, J. Org. Chem., 31, 1379
(1966).
22 P. Savignac, R. Waschbüsch, J. Carran, and A. Martinetti,
Synthesis, 7, 727 (1997).
23 M. Müller and G. Hohlneicher, J. Am. Chem. Soc., 112, 1273
(1990), and Ref. 1 cited therein.
24 K. B. Wiberg, Angew. Chem., Int. Ed. Engl., 25, 312 (1986).