вход по аккаунту


Biphenylenes with Strongly Distorted Trapezoidal Four-Membered Ring.

код для вставкиСкачать
Biphenylenes with Strongly Distorted Trapezoidal
Four-Membered Ring* *
By Fritz Vogtle,* Klaus Saitmacher. Sigrid Peyerimhoff,
Dorothee Hippe, Heinrich Puff; and Petra Biillesbach
The skeleton of biphenylene 1 has hitherto only been
described in terms of a four-membered rectangular ring
with bond dimensions 151 pm (C4a-C4b) and 143 pm
(C4a-C8b)."I We have now examined how biphenylene
might evade any additional strain"' imposed by attachment
of as short as possible a bracket at the 1,8-p0sition.~~'
problem is that distortion becomes all the more extensive
and consequential, and the synthesis all the more difficult
the shorter the length of the bridge.
In exploratory experiments we therefore tried to attach a
four-membered bridge, which, according to molecular
models exhibited little strain,[" to 1 to give the disulfide 2,
whose new ring ought to contract upon desulfurization.
3 :
L :
x = so*
4b : X
Scheme I
Reaction of the dibromo compound 6 with thiourea and
subsequent oxidation of the dithiol 7 with H 2 0 2 suffers
from the low stability of 7. In the mass spectrum of the
crude product of the synthesis, however, we surprisingly
observed signs of the existence of the monosulfide 3 with a
shorter bridge. Studies on molecular models on the other
hand, had indicated that 3 was so strained that its direct
formation in a single step was not to be expected. Encouraged by the observed formation of 3 as by-product of 2,
we allowed the dibromo compound 6 (or more simply the
mixture of 1,8- (6) and 1,5-isomer formed in its preparationC3') to react with sodium sulfide in the presence of
Cs2C0, according to the dilution principle.i41The strained
sulfide 3 was obtained in 24% yield. Oxidation with H 2 0 2
afforded the sulfone 4 in more than 90% yield.
At room temperature the 90-MHz 'H-NMR spectra of 3
and 4 show a sharp singlet for the CH,-protons (6=3.78
and 4.56, respectively, in CDC13), which is considerably
broadened in the 400-MHz 'H-NMR spectrum of 4 . The
CH, signals of both compounds split on lowering the temperature to give an AB system (Av=31 Hz, .IAB=
17 Hz
and Av=61 Hz, JAH=16 Hz, respectively; coalescence
temperatures -34" and - l O T , respectively). The ring inversion barriers (flip-flopping of the C-s-C group from
one side of the biphenylene system to the other; cf.
Scheme 1) of 3 and 4 are thus calculated to be 49 and 53
kJ/mol, respectively.[s1The small difference in the barriers
of sulfide 3 and sulfone 4 is remarkable considering a possible additional inversion of lone pairs on the sulfide sulfur, which is not possible in the case of sulfone sulfur.
The mass spectrum of the sulfone 4 shows, in addition to
the M @peak, a significant [M@- SO,] peak. This could be
an indication of the existence of the extremely strained, hitherto unknown hydrocarbon 5 . We therefore tried to obtain it by photolysis of the sulfide 3 in triethyl phosphite at
230 nm, but neither this reaction nor the pyrolysis of the
sulfone 4 at 650"C/ lo-' torr in a quartz tube afforded the
product 5 .
The result of an X-ray structure analysisi"] of the sulfide
3 is reproduced in Figure I (an X-ray structure analysis
was also carried out on 4): I n both compounds the ring
strain gives rise to very different bond lengths in the fourmembered ring. In the trapezoidally distorted four-membered ring of 4, similarly as in the case of 3, the C8a-C8b
bond (144.4 pm) is shortened and the C4a-C4b bond
(154.6 pm) is stretched; this bond is thus longer than an
isolated single bond. Remarkably, dispite this distortion
the entire biphenylene framework remains practically planar. However, the bond angles in the two benzene rings are
markedly altered by the bracketing: the ideal angle of 120"
for C8b-CILC2 is reduced to 114.5" in 4 (and even to
113.2" in 3) and that for C4a-C8b-C1 is increased to
Prof. Dr F. Vogtle, DiplLChem. K . Saitmacher
lnstitut fur Organische Chemie und Biochemie der Universitat
Gerhard-Domagk-Strasse I. D-5300 Bonn I (FRG)
Prof. Dr. S. Peyerimhoff, DiplLChem. D. Hippe
lnstitut fur Theoretische Chemie der Universitat
Wegeler-Strasse 12, D-5300 Bonn I (FRG)
Prof. Dr. H. Puff, Dr P. Biillesbach
Anorganisch-Chemisches lnstitut der Universitat
Gerhard-Domagk-Strasse I, D-5300 Bonn 1 (FRG)
We thank Dr. F. Knodi for carrying out the X-ray structure analysis of
3 -Thanks are also due to Prof. Dr W . Tliiel for supplying us with the
computer program MNDOC (correlated semiempirical calculations with
geometry optimization). The computations were carried out at the Regionales Hochschulrechenzentrum der Universitat Bonn.
0 VCH Verlag~yesellrcliafrmhH. 0-6940 Welnherm. 1987
Fig. I . Crystal structure of 3 (stereodiagram) [6]
The UV spectra of 3 (/2,,,,=359, 258 nm; c=2720 and
28 550, resp.) and
=363, 257 n m ; '= 1780 and
15 390, resp.) show measurable red shifts and significant
decreases in intensity of the long-wave bands (in ethanol)
0570-0833/87/0505-0470 S 02.50/0
Angew. Cliem. In!. Ed. Engl. 26 (1987) No. 5
compared to biphenylene (d,,,, = 278, 248 r ~ m ) ~and
' ~ dimethylbiphenylene (1,w1,8-mixture, d,,~,,=357, 254 nm).
This is an indication of perturbation of the n-electron system; analogous perturbations were observed in the transition from naphthalene to 1,8-methan0naphthalene.~'~~~
I t has been shown by MNDO calculations['o1that the
spatial structure of the molecule 3 is consistent with the
results of the X-ray crystal structure analysis. To ensure
reaching as near as possible the absolute energy minimum,
the structure was optimized in three ways: a) simultaneous
optimization of all bond lengths and angles, using approximate standard values for bond lengths and angles as starting values; b) simultaneous optimization of all bond
lengths and angles using the structural data determined by
an X-ray structure analysis as starting values; c) stepwise
optimization, involving successive optimization of the carbon skeleton of the benzene ring, then the sulfur bridge,
and then the H atoms, and repeating this procedure in the
same succession after each individual optimization. In addition, optimization was begun with a U-shaped biphenylene skeleton, to make certain that the optimization led to
a planar carbon skeleton. All three complete optimizations
afforded the values listed in Table 1 . Good agreement is
found between experimental and calculated values.
large distortion of the four-membered ring. From the existence of 3 and 4 it can be safely predicted that further
three-membered bridged biphenylenes such as 3 (X =0,
NR, CR,) with even higher ring strain and larger distortions could be synthesized.
Table I Calculated bond lengths [pml and angles ["I of 3 and 5 and, for
comparison. the values obtained by an X-ray structure analysis of 3 [a].
[ I ] J . K. Fawcett, J. Trotter, Acta Crys/al/ogr.20 (1966) 87. A. Yokozeki, C .
F. Wilcox, Jr., S. H. Bauer, J . Am. Cliem. Soc. 96 (1975) 1026.
[2] For the ring strain of biphenylene see. M. Mulin, 2. Nafurforsch.5 2 8
(1973) 478, where a value of 364 kJ/mol is quoted.
131 A longer bridge with four C atoms does not lead to any significant
changes in the bond lengths: cf. C . F. Wilcox, Jr., J. P. Uetrecht, G. D.
Grantham, K. G. Grohmann, J . Am. Chem. SOC.97 (1975) 1914: C . F.
Wilcox, Jr., D. A. Blain, J. Clardy, G. Van Duyne, R. Gleiter, M. EckertMakric, h i d . 108 (1986) 7693.
141 Cf. E. Hammerschmidt. W. Bieber. F. Vogtle, Clrenr Ber I 1 1 (1978)
2445. For further details of cesium-assistance cf. W. Kissener, F. Vogtle,
Angew. Chem. 97 (1985) 782: Anyen. Cliem. In/. Ed. Enyl. 24 (1985)
151 J. Sandstrom: Dynamic N M R Sprcrro.wip~.Academic Pres\, New York
[6] Further details of the crystal structure investigation of 3 and 4 are available on requesl from the Fachinformationszentrum Energte. Physik.
Mathematik GmbH, D-7514 Eggenstein-Leopoldshafen 2 (FRG) on
quoting the depository number CSD-52236, the names of the authors,
and the journal citation.
171 UV-Atlas of Organic Compounds, Vol. 2, Verlag Chemie, Weinheimi
Butterworths, London 1966.
[S] Cf. R. J. Bailey, H. Shechter, J Am C/rem. Soc. Y6 (1974) X I 16.
[9] The photoelectron spectrum of 3 can, as expected, be interpreted as a
superposition of the biphenylene spectrum and a sulfide band. We
thank Prof. Dr. E . Heilbronner. Basel (Switzerland), for collaboration.
[lo] M. J. S. Dewar, W. Thiel, J. Am. Chem. SOC.YY (1977) 4899. 4907.
3 (calc.)
5 (calc.)
3 (exp.)
1 I6
134.2- 135.6
138.2- 138.4
135.5- 136.4
140.9- 141.1
C 1 -CZ
C9-c I I
C8b-C I-C2
I I6
120.3- I 2 1.2
123.6- 124.4
115.2-1 15.4
127. I - 128.0
91.3- 92.3
[a] Deviations between the values calculated: I pm and 1-3", resp.; mean
deviations of the experimental values: 1-3 pm and 1-2", resp.
The hypothetical hydrocarbon skeleton 5 was also calculated in this way (Table l). A further shortening of the
C8a-C8b bond and a further lengthening of the C4a-C4b
bond are accompanied by a n even greater distortion of the
benzene ring (angles between 107 and 139"). According to
calculations, the bond between the two "bracket" carbon
atoms C9 and C11 is unusually long (161 pm).
If the enthalpy of formation AH,, is used as a measure of
the stability (or the ring strain), it follows from the calculated values AH,,=395 kJ/mol for biphenylene 112' and
43 1 kJ/mol for 3 that 5 (with 586 kJ/mol) is a considerably
less stable system. On the basis of these data it is easy to
comprehend that all previous attempts to prepare 5 met
without success.
From the calculated atomic charges of the molecules 1,
3 and 5 it follows that the charge distribution does not
essentially alter upon bracketing 1 to give 3.
3 and 4 are the shortest bracketed biphenylene derivatives known so far. The comparatively slight additional
ring strain introduced thereby results in a surprisingly
Airyen,. Chenr. I t i f . Ed. Engl. 26 11987) No. 8
2-Thia[3]( 1,S)biphenylenophane 3 : A mixture of benzene (1.6 L), elhanol
(1.2 L) and cesium carbonate (3.00 g, 9.20 mmol) 141 contained in a 4 L threenecked flask was heated to boiling and treated dropwise and synchronously
under N? for 13 h with solutions of the 56:44 mixture (formed during the
preparation [3]) of 1.8- and 1,5-bis(bromomethyl)biphenylene (2.88 g,
8.59 mmol) [corresponding to ca. 1.60 g (4.80 mmol of pure (611 in benzene
(250 mL) and NalS.9HZ0 (1.91 g, 8.60mmol) in ethanol (250mL). After
cooling, the yellow solution was evaporated down and the insoluhle cesium
compounds removed by filtration through a frit The crude product was separated by column chromatography (SO2, cyclohexaneichloroform). Aside
from unchanged starting compound, a dimeric product (C2XH21,S?)
and the
disulfide compound 2 could be detected mass spectrometrically. The desired
monosulfide 3 (240mg, 24%) was obtained as yellow crystals. m.p. 142.1 4 4 T (from n-hexaneichloroform 2 : I).-90-MHz 'H-NMR (CDCI,, TMS
int.): 6=3.78 (s, 4 H , CHZ), 6.5-6.7 (m, 6 H , aromatic H).
4 . M.p. 233°C (from acetone/cyclohexane I : I ) : M , =242 0402, high resolution MS: 242.0396.
Received: December 22, 1986:
revised: February 18, 1987 [Z 2020 IE]
German version: Angew. Chem. 99 ( 1987) 459
Dihydrocyclobutafuran" *
By Norbert Miinzel and Armin Schweig*
The thermodynamics and kinetics of the equilibrium
gas-phase reaction (a) of o-quinodimethane 1 and dihy-
Prof. Dr. A. Schweig, DipLChem. N. Miinzel
Fachbereich Physikalische Chemie der Universitat
Hans-Meerwein-Strasse, D-3550 Marburg (FRG)
Theory and Application of Photoelectron Spectroscopy, Part I I I. This
work was supported by the Deutsche Forschungsgemeinschaft and the
Fonds der Chemischen 1ndustrie.-Part 110: F. Diehl, A. Schweig, Angew. Chem. 99 (1987) 348: Angew Chem. In,. Ed. Engl. 26 (1987) 343.
0 VCH Verlag.~gesellsc/rafrmbH. 0-6940 Weinheim. 1987
057/~-/1R33/87/0505-/J471$ 02.50/0
47 I
Без категории
Размер файла
247 Кб
four, membered, ring, strongly, biphenylene, distorted, trapezoidal
Пожаловаться на содержимое документа