close

Вход

Забыли?

вход по аккаунту

?

Dicyclopropa[a c]naphthalene a Compound Capable of Existence.

код для вставкиСкачать
DicyclopropaIa,c]naphthalene :
a Compound Capable of Existence?**
By Udo H. Brinker.* Hartmut Wiister. and Gerhard Maas
Cycloproparenes have been the subject of intensive
study for almost 25 years,"] yet dicycloproparenes, which
have two three-membered rings fused to the same aromatic
ring, are thus far unknown. All cycloproparenes have a
high strain energy; nevertheless, their kinetic stability
strongly depends on their structure. Linearly coupled cycloproparenes such as 1[*"-"]and 2[2'1 are generally more
stable and also easier to prepare than, e.g., the angular species 3''' and 4.141Thus, 3 explodes on melting,13.4a1while
solutions of 4 decompose on prolonged standing, even at
-60°C.
1
2
3
4
5
6
7
8
w
The dicyclopropa[b,g]naphthalene SCs1synthesized by
Vogel et al. is the only dicycloproparene to have been isolated so far. Its strain energy has been determined as 166
kcal . mol- ' (lower limit);'61 it is sensitive to shock and explodes on melting.
Despite this there has been no lack of effort in attempting to prepare the more highly strained dicycloproparenes,
in which one aromatic ring and two three-membered rings
are annelated. Attempts to generate the parent compounds, the dicyclopropaIa.d)benzene 6[7"1and dicyclopropa[a,c]benzene 7,['] have thus far failed. The species 8,
benzoannelated 7, is expected to have a higher strain energy than 5, in which the three-membered rings are located
at the maximum distance from each other. We report here
on the preliminary results of our attempts to generate and
trap the species 8.
An ideal starting compound for the generation of 8 is
the benzoannelated r- 1,t-2-dibromotricyclo[5. I .0.0'.4]oct-5ene lo''', which is obtained by 1,6-dehalogenati0n'~.'~
of the
(&)-bisdibromocarbene adduct 9 with methyllithium at
-78°C (yield 60%). The dicycloproparene 8a could be
formed by twofold HBr-elimination from 10, whereby the
benzylic, tertiary protons (H-4 and H-7) should be attacked by a suitable base. Besides this there is the possibility of a competing 1,2-elimination taking place,'41initiated
by cleavage of one of the geminal cyclopropane ring protons. This process leads to an "exocyclic" double bond in
the three-membered ring, which is not conjugatively
bonded to the benzene ring (see 11).
Upon reaction with potassium tert-butoxide (2.4 mol.
equiv.) at 20°C in T H F / D M S O ( 1 : l), 10 is already no
longer detectable by gas chromatography after a few minutes. N o significant signals could be observed in the temperature range from + 2 0 to -70°C during the ' H - N M R
and I3C-NMR spectroscopic monitoring of the reaction of
12
11
hl
/
.
& B ~ ~
MeLi
Brz
-7@C
rB
&
-.-
H;
5
10
(?I-9
3z
Br
fBuOK
DPIBF
_c
Ha
4
Br
13
-
&
.
Br
14
\
Ph
17
15
[*I
Prof. Dr. U. H. Brinker, D i p L C h e m . H. Wiister
Fakultat fur Chemie d e r Universitat
Universitatsstr. 150, D-4630 Bochum-Querenburg (FRG)
Prof. Dr. G. Maas ['I
Fachbereich Chemie d e r Universitat
Erwin-Schrodinger-Str., D-6750 Kaiserslautern ( F R G )
['I
[**I
X-ray structure analyses.
This work was supported by the Fonds d e r Chemischen lndusrrie a n d
by Degussa AG. We thank Dr. W.Dierrich and Dip].-Chem. H . Kiihne
for the N M R spectra of 12, 15, a n d 16 and Herr K . Gomann for assistance with the synthesis of 10.
Angens Chem. In,. Ed. Engl. 26 11987) No. 6
16
10 (2.4 mol. equiv. KOtBu, 2.4 mol. equiv. 18-crown[6],
[DJTHF).
Of the two possible KekulC forms of 8, only 8a can be
formed directly from 10 by twofold 1,2-elimination. 8a
could be the energetically
stable form, since it contains an intact benzene ring and two c!fclopropene
Attempts to trap the diene unit in 8a with 4-phenyl-1,2,4-
0 VCH VerlngsgesellschaJ? mbH. 0-6940 Weinheim. 1987
0570-0833/87/0606-1)577
S 02.50/0
577
triazol-3,5-dione (PTAD) as dienophile met without success, since KOtBu destroys the trapping agent.["'
In view of the expected instability of 8 it was considered
advisable from the very start to use the reactive 1,3-diphenylisobenzofuran (DPIBF) as diene." 'I Reaction of 10
with DPIBF in the molar ratio 1 :9 (2.4 mol. equiv. KOtBu, T H F :DMSO = 1 : 1, 20°C) led to the formation of
three crystalline bisadducts (yield 8l%), which could be
separated by HPLC (silica gel Si 60-5, pentane:
CH,CI, = 6 :4). The bisadduct 12 (yield 28%) is formed via
twofold Diels-Alder addition of DPIBF to 11 or one addition to 13 with formation of 14, which after HBr-abstraction and creation of an "exocyclic" double bond (cf. 11) is
attacked again by DPIBF. The chemical shifts of H-3 and
H-4 in 12 are in good agreement with those of the corresponding protons of the adduct of furan to C-1 and C - l a
of 4.I4.I2]The crystal structure a n a l y ~ i s ~can
' ~ ]explain the
observed high-field shift of two aromatic protons at
6=5.25 and 5.6.
Since the number of '3C-resonance lines is only half as
many as the maximum possible lines, the main product 15
(yield 44%) must be a symmetrical compound. The threemembered rings are oriented in the ~ n t i - p o s i t i o n , ~ ' ~ ]
whereas the cyclopropane ring and 0-atom of a five-membered ring are each syn-oriented. The signals of the protons on C-3 and C-8 (6= -0.5), which point towards the
benzene ring, are shifted upfield. In contrast, the protons
oriented in the direction of the oxygen atoms (6=2.75) are
shifted markedly downfield. The overall result is that 15 is
formed by twofold endo-addition of the trapping agent.
In contrast, 16 (yield 9%) can be formed by an endo- and
an exo-oriented [4 2]-cycloaddition. The characteristically
large difference in the chemical shifts of the protons on
C-8 observed in the case of 12 and 15 is not found in 16.
As indicated by the X-ray structure analy~is,"~]
both protons on C-8 are similarly shifted noticeably downfield by
the 0-atoms. Altogether, the Diels-Alder reactions of
DPIBF with the double bonds generated from 10 or 13
and 14 with KOtBu with formation of the bisadducts 12,
15, and 16 proceed with strongly preferred endo-attack
+
(94%).
The crucial question is, whether dicyclopropa[a,c]naphthalene 8 is in fact formed as an intermediate in the reaction of 10 with KOtBu, or whether o r not two cyclopropene systems generated by consecutive I ,2-eliminations are
each trapped directly by DPIBF as bisadducts 15 and 16.
A concerted twofold HBr-elimination in 10, on the other
hand, seems unlikely.
As evidenced by the formation of 12 in 28% yield, not
only is a conjugated "endocyclic" double bond formed but
also, to the extent of ca. 17% (based on all 1,2-eliminations), the strained "exocyclic" double bond of the cyclop r ~ p e n e s . "The
~ ~ fact that ca. 83% of the DPIBF adducts
12, 15, and 16 result from additions to the double bonds
generated between C-l and C-7 and/or between C-2 and
C-4, proves the strongly preferred direction of elimination
with cleavage of benzylic hydrogen atoms and formation
of conjugated double bonds.1i51Under the standard conditions with 2.4 mol. equiv. of base, formation of the monoadduct 14, as is expected in the addition of DPIBF to the
monoelimination product 13, could not be detected. When
only 1.1 mol. equiv. of the base was used, 45% of the monoadduct 14 could be isolated along with 9% of the bisadducts 12, 15, and 16 (ratio ca. I :2.5 : 1). In the reaction of
14 with 1.2 mol. equiv. of base and DPIBF under the conditions employed for the reaction of 10, three bisadducts
(24% of 12, 48% of 15, and 10% of 16) were isolated (ratio
578
0 VCH Verlagsge~ellschaflmbH. 0-6940 Weinheim.I987
2.4 :4.8 : 1). The comparable ratios of the bisadducts on
starting from 10 and 14 suggests that the monoadduct 14
should also be formed as intermediate in the reaction of 10
with 2.4 mol. equiv. of base.
The annelation of a cyclopropane ring to the benzene
nucleus effects a strain increment of ca. 14 kcal .mol - I in
excess of the sum of the E, of the structural units benzene
and cyclopropene [difference of the strain energies E, of 1
(68 kcal . mol
and of cyclopropene (53.7
kcal . mol - I)]. Accordingly, for dicyclopropa[a,c]naphthalene 8 an increase in strain is expected which exceeds the
strain energies of the two cyclopropene units contained in
8 by at least 28 kcal .mol I. This large increase in strain
energy obviously cannot be compensated by the gain in
aromatization energy. When starting from 10 in the presence of KOtBu, instead of the formation of 8a or of the
aromatic system 8, stepwise 1,2-eliminations and additions
of DPIBF apparently take place.
Received: January 28, 1987:
revised: March 13, 1987 [Z 2074 IE]
German version: Angew. Chem. 99 (1987) 585
[I] a) B. Halton, lnd. Eng. Chem. Prod. Res. Deu. 19 (1980) 349; b) W. E.
Billups, Acc. Chem. Res. I 1 (1978) 245.
121 a) E. Vogel. W. Grimme, S. Korte, Tetrahedron Lett. 1965. 3625; b) E.
Vogel, S. Korte, W. Grimme, H. Giinther, Angew. Chem. 80 (1968) 279;
Angew. Chem. Int. Ed. Engl. 7 (1968) 289; c ) B. Halton, P. J. Mifsom,
Chem. Commun. 1971. 814; d) W. E. Billups, A. J. Blakeney, W. Y .
Chow,ibid. 1971. 1461; e) W. E. Billups, W. Y. Chow, J. Am. Chem. SOC.
95 (1973) 4099.
131 S. Tanimoto, R. Schafer, J. Ippen, E. Vogel, Angew. Chem. 88 (1976)
643; Angew. Chem. I n t . Ed. Engl. 15 (1976) 613.
[4] a) B. Halton, B. R. Dent, S. Bohm, D. L. Officer, H. Schmickler, F.
Schophoff, E. Vogel. J . Am. Chem. Soc. 107 (1985) 7175; b) P. Miiller,
H. C. Nguyen Thi, J. Pfyffer, Helu. Chrm. Acra 69 (1986) 855.
IS] J. Ippen, E Vogel, Angew. Chem. 86 (1974) 780: Angew. Chem. lnr. Ed.
Engl. 13 (1974) 736.
(61 F. D. Rossini, W. E. Billups. unpublished, cited in W. E. Billups, Arc.
Chem. Res. I 1 (1978) 245.
[7] a) M. G. Banwell, B. Halton, Ausl. J. Chem. 32 (1979) 849, 2689: b) E.
Vogel, W. Piittmann, W. Duchatsch, T. Schieb, H. Schmickler, J. Lex,
Angew Chem. 98 (1986) 727; Angew. Chem. In,. Ed. Engl. 25 (1986)
720.
[S] a) U. H. Brinker, H. Wiister, G. Maas, J. Chem. Soc. Chem. Commun.
1985. 1812; b) J . Hohn, P. Weyerstahl, Chem. Ber. 116 (1983) 808.
191 Concerning bond fixation in cyclopropabenzene see: Y. Apeloig, D.
Arad, J. Am. Chem. SOC.108 (1986) 3241.
[lo] The addition of the dienophile PTAD to the diene moiety in 8a should
lead to a markedly nonplanar double bond between the three-membered
rings.
[ I I ] I n order to check to what extent cycloproparenes actually undergo cycloadditions with DPIBF at 20°C and normal pressure we have used 1
as a model species. Whereas 1 does not react with the less reactive 1,4diphenyl- 13butadiene under these conditions (B. Halton, unpublished
results, cited in [la]), with DPIBF it forms the endo-adduct 17. The stereochemical assignment is based on the deshielding of the proton in the
anti-position to the norcaradiene unit by the oxygen atom (6=3.15). Accordingly a [2+4]- o r [6+4]-cycloaddition to C-l and C-6 i n 1 could
have taken place.
[I21 T. T. Coburn, W. M. Jones, J. Am. Chem. SOC.96 (1974) 5218.
1131 The structures of 12 and 16 have been confirmed by X-ray structure
analyses.
1141 On the basis of the spectroscopic data it cannot be completely ruled out
that the cyclopropane rings in 15 are in thesyn-position. The reaction of
13 with DPIBF to give 14 is an endo-addition. However, a second endoaddition of DPIBF-to the "conjugated" cyclopropene resulting from
reaction of 14 with the base-requires a sterically unfavored transition
state. This is not the case in the formation of 15 with anti-oriented threemembered rings.
[I51 DPIBF-adducts, derived from a compound with two strained "exocyclic" double bonds, were not found.
1161 W. E. Billups, W. Y. Chow, K. H. Leavell, E. S . Lewis, J. L. Margrave,
R. L. Sass, J. J. Shieh, P. G . Werness, J. L. Wood, J . Am. Chem. SOC.95
(1973) 7878.
0570-0833/87/0606-0578 $ 02.50/0
Angew. Chem. Int. Ed. Engl. 26 11987) No. 6
Документ
Категория
Без категории
Просмотров
2
Размер файла
239 Кб
Теги
naphthalene, compounds, dicyclopropa, existencia, capable
1/--страниц
Пожаловаться на содержимое документа