вход по аккаунту


Are Lithiosulfones Configurationally Stable.

код для вставкиСкачать
covalent stabilization, thus being the dominant interaction.
The similarity of the structures 1-3 and of their interactions
indicates that NADPH and FAD are arranged in a nearly
ideal fashion for a hydride transfer to N-5 of the isoalloxazine ring in the enzyme.
In the second step, FADH' transfers reduction equivalents to the disulfide bridge Cys58:Cys63 (cf. 5, bottom). In
this process, the direct nucleophilic attack of the reduced
isoalloxazine ring offers an attractive possibility corresponding to the usual proposal of nucleophilic attack on FAD.[91
In such an attack, the nucleophile forms a covalent bond to
C-4a of the isoalloxazine. Crystal structure analysis and calculation supports such a mechanism. Especially interesting is
the observed unusual conformation of the disulfide bridge,[31
which, in the given fixed position relative to the isoalloxazine
ring, leads to an exact orientation of the antibonding L U Mo
of the bridge toward C-4a, as is shown by the scaled reproduction of the crystal structure of 5. C-4a of the isoalloxazine
ring and the S atoms of the disulfide bridge lie next to each
other in a linear fashion, and the energetically nearly degenreerate frontier orbitals (HOM0,,,,e-LUM0,,,,8:Cys63)
sult in a strong covalent interaction between C-4a and the
sulfur atom of Cys63. Structure 5 serves to emphasize the
favorable orientation of the orbitals for back-side attack of
C-4a on the disulfide bond. The mechanism can be described
as a nucleophilic cleavage of the SS bridge with formation of
a covalent bond between C-4a of the isoalloxazine and the S
atom of Cys63. Transfer of the proton from N-5 to the S
atom of C y ~ 5 and
8 ~ fragmentation by intramolecular nucleophilic substitution to give Cys63" should follow. It is interesting to note that the MNDO PM3 calculation for the
interaction of FAD with Cys63@, the relative position of
which is known from X-ray data, gives a stabilization of
- 2.6 kcal mol-' and shows no repulsion, although this is
usually observed between molecules with closed valence
shells. The calculation apparently confirms the spectroscopically established formation of a CT complex['. 21 between
FAD and Cys63'.
The MNDO PM3 calculations and the perturbation analysis of the structure of the active site of glutathione reductase
show that, in the enzyme, the reaction partners NADPH and
FAD, as well as FADH' and Cys58:Cys63, are positioned in
such a way that they are localized near the transition state
along the reaction coordinates of the hydride transfer and
the nucleophilic cleavage of the disulfide bridge, respectively.
This positioning eliminates the decrease in entropy required
for this reaction and is in agreement with proposals concerning enzyme-catalyzed reactions in general. Furthermore,
other mechanisms, such as one-electron transfers, become
improbable. Further investigations, which also consider the
reduction of glutathione and include some important amino
acids in the vicinity of the active center, are in progress.
Rcccived: March 3, 1989;
supplemented: May 11, 1989 [Z 3213 IE]
German version: Angew. Chem. fOf(1989) 1056
[I] C. H. Williams, Jr. in P. D. Boyer (Ed.): The Enzymes, Vol. 13,Academic
Press, New York 1976, p. 89.
[2] E. F. Pai, G. E. Schulz, 1 Biol. Chem. 258 (1983) 1752; E. F.Paj, P. A.
Karplus, G. E. Schulz, Biochemistry 27 (1988) 4465; P. A. Karplus, E. F
Pdi. G. E. Schulz, Eur. J. Biochem. 178 (1989) 693.
I31 P. A. Karplus, G. E. Schulz, J Mo/. Biol. 195 (1987) 701; P. A. Karplus.
G. E. Schulz. J. Mol. Biol., in press.
[41 J. J. P. Stewart, J. Comput. Chem., in press. We thank Dr. 7: Clark, Universitiit Erlangen-Niirnberg. for providing us with a program containing the
MNDO PM3 parameters.
A n g c ~ Chem.
In[. Ed. Engi. 28 ( I989) No. 8
[S] R. Sustmann, W. Sicking, Chem. Ber. 120 (1987) 1323; R Sustmann, P.
Daute, R. Sauer, W. Sicking, Tetruhedron Lett. 29 (1988) 4699, and references cited therein.
0.Tapia, R. Cardenas, J. Andres, P. Colonna-Cesari,J Am. Chem. SOC.110
(1988) 4046.
Y-D. Wu, K. N. Houk, J. Am. Chem. SOC.109 (1987) 906.
J. W. Verhoeven, W. van Gerresheim, F. M. Martens, S. M. van der Kerk,
Teiruhedron 42 (1986) 915.
C. Walsh, Acc. Chem. Res. 13 (1980) 148.
Are Lithiosulfones Configurationally Stable? **
By Hans-Joachim Gais,* Gunther Hellmann, Harald Giinther,
Fernando Lopez, Hans J. Lindner and Sigmar Braun
Dedicated to Professor Christoph Riichardt on the occasion of
his 60th birthday
According to crystal structure analyses, lithiosulfones 1
have chiral anions even when their a-C atom, bearing different groups, is coordinated in a planar rather than in a tetrahedral fashion.['] In this respect they differ fundamentally
from lithium en01ates.L~~
This is due to a favored C,S conformation in which the two Li-bound 0 atoms are arranged
gauche to the lone pair on the a-C atom (cf. Fig. 1 and the
formula shown). Earlier H/D-exchange experiments on optically active sulfones 2 have already provided compelling evidence for the chirality of counterion-free a-sulfonyl carbanions and shown that hindered C,-S rotation, not C,
inversion, is responsible for their ob~ervability.[~~
Ab initio
calculations came to a similar conclusion.r51According to
these results negative hyperconjugation (n,-o,*,,) and Coulombic interaction are the dominant mechanisms of stabilization. How stable, however, are the configurations of the
lithiosulfones 1 ? This important question has remained
unanswered so far. In particular, no evidence has been provided for the chirality of 1 in solution.16]We were also especially interested in whether optically stable lithiosulfones 1
can be generated enantioselectively from optically active sulfones 2.
When the metalation of optically active phenyl sulfone 2a
with nBuLi in T H F was followed by polarimetry, only rapid
[*] Prof. Dr. H.-J. Gais, DipLIng. G. Hellmann
Institut fur Organische Chemie und Biochemie der Universitat
Alberfstrasse 21, D-7800 Freiburg (FRG)
Prof. Dr. H. Giinther, Dr. F. Lopez
Fachbereich 8, Organische Chemie I1 der Universitat
Postfach 10 1240, D-5900 Siegen (FRG)
Prof. Dr. H. J. Lindner, Dr. S. Braun
Institut fur Organische Chemie der Technischen Hochschule
Petersenstrasse 22, D-6100 Darmstadt (FRG)
This work was supported by the Deutsche Forschungsgemeinschaft, the
Fonds der Chemischen Industrie, the Wissenschaftliche Gesellschaft in
We thank
Freiburg, and the Alexander von Humboldt Foundation ( E L).
Dr. W Buuer for the cryoscopic measurements, Dr. A . Powell and Dr. H .
Puulns for the crystallographic data sets, and Dr. E. KeNer for a version of
YCH Verlagsgesellst.huft mbH. 0-6940 Weinheim. 1989
S 02.50/0
racemization with formation of rue-1 a was observed, even at
- 80°C. 'H DNMR measurements on the benzyl-substituted lithio(phenyl)sulfones rue-1 b and rue-1 c, in which the
exchange of nonequivalent methylene protons was observed
in [D,]THF, gave the following results: roc-1 b, the a-C atom
of which should be coordinated in a planar fashion,['] has an
enantiomerization barrier AG:15 of 9.6 i 0.2 kcal mol-'
(c = 1.8 x l o - ' rnol L-'); rue-lc, the a-C atom of which
should be coordinated, by contrast, in a tetrahedral
fashion,['] likewise has a AG:15 value of only 9.7
0.2 kcal mol-' (c = 2.8 x l o - ' rnol L-').['] Over a concentration range of 7 x 1 0 - ' to 36xIO-'molL-', the A G *
value for ruc-lb does not change, but it increases to
10.1 k 0.2 kcal mol- ' in the presence of four equivalents of
hexamethylphosphoric triamide.I81
Trifluoromethyl sulfones (tr~ji'onrs)are ca. 10 pK, units
more acidic than phenyl s u l f ~ n e s . We
[ ~ ~found that lithiotriflones are also appreciably more configuratively stable than
lithio(pheny1)sulfones. Metalation of the optically active triflone (+)-2d["] with nBuLi in THF at - 95 "C resulted in
formation of the optically active lithiotriflone (-)-Id with
an extrapolated half-life of 30 d at -78 "C. The time dependence of the racemization of (-)-1 d was followed polarimetrically at low temperatures for a concentration range of
to 120 x
rnol L-'. It obeys first-order kinetics
with a racemization barrier AG:,, of 17.2 f 0.1 kcal mol-'
s-', t , , , = 12.3 min).["] A A H * value
(k,,, = 9 . 4 0 ~
of 16.7 k 0.3 kcalmol-'
and a A S * value of
- 1.9 f 1 .I cal K - ' mol-' were derived from the temperature dependence (-44 to - 24 "C) of k,,,. A linear relation
between a and t was also obtained for the racemization of
(-)-1 d in T H F in the presence of four equivalents of 1,3dimethyltetrahydropyrimidine-2(1I j ) - ~ n e . [In
~ ~ this case,
the lifetime of (-)-1 d (fl,, = 42 min at - 34 "C) is longer.
By 'H DNMR measurements on rue-ld, the a-C atom of
which should be tetrahedrally coordinated,['] an enantiomerization barrier AG&3 of 17.8 f 0.2 kcal mol-'
(c = 3.2 x lo-' rnol L-') was found following the exchange
of methylene protons in [D,,Jdiglyme.[81 For the enantiomerization of vuc-le, the a-C atom of which has been
shown to be coordinated in a planar fashion (cf. Fig. I), the
NMR measurements in [D,]THF gave a AG:33 value of
16.0 f 0.2 kcal mol-' (c = 1.9 x l o - ' rnol L-')."] According to cryoscopic measurements,"'"' 1 e is monomeric in
T H F at - 103°C and concentrations of 7.1 x lo-' and
13.6 x IO-'mol L-l,[lZb'C1and,based onNMRanalysis,associated with two molecules of T H E The results of NMR
investigations on (-)-1 d and rue-1 d ([D,]THF, - 80 "C,
c = 3.1 x l o - ' mol L-') are in agreement with these findings. Under these conditions, chiral and achiral dimers
would have to be present in addition to chiral monomers.
There are no differences in the positions of either the ' H or
the 13CNMR signals, and only one species is observed. The
appreciably higher acidity of 2d,e and the greater configurative stability of l d , e are in accord with the stabilization of
a-sulfonyl carbanions by negative hyperconjugation and
Coulombic interaction.". ' 31 Hindered C,-S rotation is the
reason for the observability of chirality in 1e. This should
also be true for 1 a-d, whereby the a-C atoms in 1c-d might
be coordinated in a tetrahedral rather than a planar fashion.121
The metalation of (R)-2d and the protonation of (-)-1 d
occur enantioselectively and with retention of configuration!
Reaction of ( - ) - l d (c = 7.3 x lo-' rnol L-I), generated
from (R)-2d['01 and one equivalent of nBuLi at - 90 "C in
THF, with CF,COOH in T H F at - 105°C afforded (R)-2d
(93 %) with [a]:& = + 10.9 (c = 2.8, THF).
(c> VCH VerlagsgesellschafimhH, 0-6940 Weinheim, 1989
Are the differences in the stability of deprotonated and
lithiated triflones and phenyl sulfones reflexted in C,-S
bonds of different length? To answer this question, crystal
structure analyses of (1 e . TMEDA),['4"] and 2e[14'] were
carried out.[''I Single crystals of ( l e . TMEDA), were obtained in the usual way by metalation of 2 e with nBuLi
in N,N,N',N'-tetramethylethylenediamine (TMEDA). (1 e .
TMEDA), crystallizes as a typically achiral dimer"] with
(M) or (P)configuration['61 of the two anions, the a-C
atoms of which are coordinated in a planar fashion (Fig. 1).
Fig. 1. Molecular structure of one of the two crystallographically independent
dimers of (1 e . TMEDA), [17]. Selected bond lengths [A] and angles ["Ifor both
dimers: Sl-01 1.455(4), S1-02 I .446(5), C1-Sl 1.856(6), C2-S1 1.608(6),
C2-C3 1.455(9), Lil-01 1.920(9), Lil-02 1.877(9), Li1-Nl 2,13119).
119.2(5) C2-Sl-Cl
114.9(3); Cl-Sl-C2-C3
- 80.1,
Cl-Sl-C2-C9 96.7. Sl-01 1.445(4), S1-02 1.459(4), C1-S1 1.823(8),
C2-Sl 1.611(6). C2-C3 1.480(9), Li2-01 1.910(12), Li2-02 1.907(9),
LIZ-Nl 2.132(14), Li2-NZ 2.130(11);C3-C2-S1 123.0(4), C9-C2-S1 116.5(5),
112.3(3); Cl-Sl-C2-C3
Cl-Sl-C2-C9 - 92.6.
Apart from the N atoms of the TMEDA molecules, only the
0 atoms of the sulfonyl groups are bound to the Li atoms.['81
Bonds do not exist between the Li and the a-C atoms. The
anions have the characteristic C,S conformation;['] however, their C,-S bond is shortened by l l % to l .609 A (average value) compared with that in 2e. It is thus shorter than
that in (1 a . diglyme), (1.652 A),["] the a-C atoms of which
are also coordinated in a planar fashion, and of the same
So far,
length as that in [(CH,SO,Ph)Li . TMEDA],
shortenings of the C,S bond in metalated sulfones of only
ca. 8 YOhave been found.['] The 0-Li distances are of roughly the same length as those in all other lithiosulfones, while
the S-0 distances are slightly increased. Noteworthy is the
fact that the S1-C1 bonds in (le.TMEDA), (average
1.839 A) are no longer than the corresponding bond in 2 e
(average 1.853 A).
Finally, it was important to determine whether the Li
atom of monomeric 1 e in solution occupies the same position (0-associated) as the Li atoms in (1 e . TMEDA), in the
crystal. To this end, 6Li,'H heteronuclear NOE experiments[201were performed on 1e at 25 oC.lz'lFor these experiments, a solution of crystalline (1 e . TMEDA), prepared
with 95% 6Li in [D,]THF (c = 3 x lo-' mol L - I ) was used.
On the one hand, 2 D NMR experiments using the HOESY
z 2 1 allow one to observe intense correlapulse sequence[20"*
tion peaks for the CH, and CH, protons of TMEDA, which
indicate that, even in [D,]THF solution, the Li atom is
chelated by TMEDA. On the other hand, correlation peaks
Angen. Chem. h i . Ed. Engl. 28 (1989) No. 8
for 1 1-H/15-H (6 = 7.38) as well as for 4-H/8-H and/or 12H/14-H were observed; the signals of the latter (6 = 7.21 and
7.1 4, respectively) lie very close to each other.[231Against
expectations, however, no interactions with the diastereotopic CH, protons could be obtained in this experiment,
despite variation of the mixing time over a range of 2 to 5 s.
Therefore, one-dimensional NOE difference experiments
were also carried out and, upon irradiation of the resonance
frequencies of the CH, protons, the expected intensity enhancements were observed. Moreover, the Li,H interactions
already appearing in the HOESY spectrum were revealed in
the corresponding 1 D NOE difference measurements
(Fig. 2). Furthermore, despite the small difference in shifts,
Review on crystal structures of lithiosulfones. G. Boche. A n g w . Chem.
101 (1989) 286. Angen. Chem. Inr. Ed. Engl. 28 (1989) 277.
a) [(Me,CSO,Ph)Li diglyme], [2b] and [(CH, =CHC(H)SO,Ph)LI . diglymejz [2c] have tetrahedrally coordinated a-C atoms and [(PhC(H)S02Me)Li. TMEDA], [l], [(PhC(Me)SO,Ph)Li . diglyme],. [2b] and
[(Me,SiC(H)SO,Ph)Li . TMEDA], [I] have planar-coordinated %-C
atoms; b) H.-J. Gals, J. Vollhardt, G. Hellmann. H. Paulus. H. J. Lindner.
Tetrahedron Lett. 29 (1988) 1259; c) H.-J. Gais, J. Vollhardt. H. J. Lindner,
Angen. Chem. 98 (1986) 916; Angeu. Chem. Int. Ed. Engl. 25 (1986) 939.
Review on the crystal structures of lithium enolates: D. Seebach. Angrw.
Chcm. 100 (1988) 1685; Angew. Chem. Inl. Ed. Engl. 27 (1988) 1624.
E. J. Corey, T. H. Lowry, Telruhedron Left. 1965,803; J. N. Roitman. D. J.
Cram, J. Am. Chrm. Soc. 93 (1971) 2225.
a) S. Wolfe, Stud. Org. Cliem. (Amstcvdam) 19 (1985) 133; b) D. A. Bors.
A. Streitwieser. Jr., J. Am. Chem. Soc. 108 (1986) 1397.
Cram el al. (A. Ratajczak, F. A. L. Anet. D. J. Cram, J. Am. Chem. Soc. 89
(1967) 2072) showed by 'H NMR spectroscopy that lithio(2.2-dimethylcyclopropyl) phenyl sulfone is chiral and found an enantiomerization barrier
AG& of 18.0 kcal mol-'.
Estimated from the coalescence point as described by R. J. Kurland, M. B.
Rubin, M. B. Wise, J. Chem. Phys. 40 (1964) 2426.
Result of a line-shape calculation as described by S. Alexander, J. Chrm.
Pllys. 37 (1962) 974.
F. G. Bordwell. Ace. Chem. Rrs. 21 (1988) 456.
(R)-2d ([z]::~
14.5 (c = 2.8, THF) was prepared from (R)PhCH,CH(Me)SH
- 13.5 (neat [lob]) by reaction with CF,Br
(NHJTHF, - 80°C. hv) according to Ignot'rv et al. [loc] to give ( R ) PhCH,CH(Me)SCF, ([a]& = - 3.2 (c = 5.0, THF). 72%) followed hy
oxidation (H,O,, H,WO,, glacial acetic acid, 92%) according to
Huze/dinr el al. [lOd]; b) C. L. Arcus, P. A. Hallgarten, J. Am. Chem. Soc.
1956, 1987; c) N. V. Ignat'ev, V. N. Boiko, L. M. Yagupol'skii, J. Org.
Chrm. ( U S S R ) 21 (1985) 592; d) R. Hazeldine, R. B. Rigby. A. E. Tipping, J Chem. Soc. Perkin Trans. i 1973. 676.
A first order reaction was verified graphically over at least four half-lives.
The calculation of k,,, (Krdc= 2 k,,,,) as well as of the activation parameters (Eyring procedure) was carried out with a kinetic program.
a) W. Bauer, D. Seebach, Helo. Chim. Acta 67 (1984) 1972; b) W. Bauer
(Erlangen), private communication; c) [PhC(H)SO,Ph]Li is also
monomeric in
25'C and a concentration range of c = 1.35 x
mol L - ' : M. J. Kaufmann, S. Gronert. D. A. Bors. A. Streitwieser, Jr., J. Am. Chem. Soc. 109 (1987) 602.
a) M. J. Janssen in C. J. M. Stirling (Ed.): Organic Sulphur Chrmisrry,
Butterworths, London 1975, p. 19; b) Interestingly, the N-S rotation barrier in R'R2NS0,R3 for R3 = CI is appreciably higher than for R3 = Me
or Ph: W. B. Jennings, R. Spratt, J. Chem. Soc. 1970, 1418.
a) (1 e TMEDA),: space group PT, a = 10.987(6). h = 11.256(5), c =
19.556(8) A, z = 81.49(4), fi = 84.12(4). i = 76.99(4)", V = 2324.5 A3,
2 = 4,
= 1.247 gem-,, ~(Mo,.) = 1.42 cm-'. Intensity measurements at room temperature, CAD-4 diffractometer, Mo,, radiation,
graphite monochromator. 6592 measured reflections (2" < 2 0 i48").
3747 symmetry-independent reflections with F > 3 a(F). SHELX86; direct methods yielded two crystallographically independent centrosymmetric dimers which do not differ significantly. Anisotropic refinement of the
positioned H atoms, 562 variables. R = 0.084, R, = 0.087. Maximal residual electron density 0.63e/A3 near the S atoms. b) 2e: Pn2,u (Pnu2,).
a = 18.693(5), b = 10.527(3), c = 15.234(4) A, V = 2997.8 A'. Z = 8.
= 1.393 g cm-,. R = 0.041, R, = 0.056. Selected bond-length [A]
and angles (for numbering see Fig. 1): S1-01 1.42713). S1-02 1.426(3),
C1-Sl 1.850(6), C2-Sl 1.807(3), C2-C3 1.518(4); C2-S1-Cl 102.6(2),
SlA-OlA 1.425(3), SlA-O2A 1.421(3), Cl A-SlA 1.857(5), C2A-SlA
1.803(3), C2A-C3A 1.506(5); C2A-SIA-C1A 102.9(2). Further details of
the crystal Structure investigations may be obtained from the Fachinformationszentrum Karlsruhe, Gesellschaft fur wissenschaftlich-technische
Information mbH, D-7514 Eggenstein-Leopoldshafen 2 (FRG), on quoting the depository number CSD-53850, the names of the authors, and the
journal citation.
Crystal structure of Rb[CH(SO,CF,),]: K. T. Davoy. T. Gramstad, S.
Kuseby, Actu Chem. Scand. A33 (1979) 359.
V. Prelog, G. Helmchen, Angew. Chem. 94 (1982) 614; Angew. Chem. In!.
Ed. Engl. 21 (1982) 567.
E. Keller, Schakal-8R, A FORTRAN Programm,for the Gruphic Representation of Moleculur und Crys~ullographicModels, Freiburg 1988.
The crystal structure of ( l e . 2 THF), corresponds essentially to that of
( l e . TMEDA),: H. J. Lindner, H:J. Gais, G. Hellmann, unpublished.
J. Vollhardt, Angew. Chem. 9711985)865; Angew.
H:J. Gais, H. .I
Chem. Int. Ed. Engl. 24 (1985) 859.
a) W. Bauer, M. Feigel, G. Muller. P. von R. Schleyer, J. A m . Chem. Soc.
110 (1988) 6033; W. Bauer. P. von R. Schleyer, Mugn. Reson. Chem. 26
(1988) 827, and references cited therein; b) H. Gunther, D. Moskdu, P.
Bast, D. Schmalz, Angew. Chem. 99 (1987) 1242; Angew. Chem. I n t . Ed.
Engl. 27 (1987) 1212.
6Li. 'H HOESY experiments: Bruker AC 300, program HOESY (Bruker
Software): 128 increments in I , with 48 pulses each, mixing time 4 s, delay
Fig. 2. 'Li{ 'H} NOE difference spectra of 1e TMEDA (see formula for numbering) in [D,]THF at 298 K. The irradiation in the 'H NMR spectrum and the
resulting 'Li NMR difference spectra are joined by broken lines. 0 = solvent
measurement with decreasing irradiation energy showed
that only 4-H/8-H/1 l-H/lS-H, and not 12-H/14-H,
exert an NOE on the 6Li atom. The short Li-H distances
foundrz4]and the other results make a chiral structure
of the type shown in the formula for l e . TMEDA in THF
solution very probable. An (achiral) structure of this type
has already been found in an X-ray structure analysis of
[(Me,Si),C-SO,Ph]K . [I 8 ] c r 0 w n - 6 [ ~and
~ ~ in theoretical
investigations of (CH,S0,Me)Li.[5bl
Received: March 8, 1989;
Supplemented: April 27, 1989 [Z 3220 IE]
German version: Angel<,. Chem. 101 (1989) 1061
Angew. Chem. Int. Ed. Engl. 28 11989) No.8
Verlugsgesellv hafr mhH, 0-6940 Wernheim.1989
time 4 s, data matrix 256 ( t i ) x 512 ( I n ) , Gauss filter. 'Li('H] NOE difference spectra: program NOEDIFF: irradiation time 3 s, power 35 L, delay
time 7 S, 160 pulses per experiment.
C. Yu, G . C. Levy, J. Am. Chem. Sbc. 106 (1984) 6533.
The assignment of the phenyl group signals in the ' H N M R spectrum was
achieved by 'H,'H-COSY and "C,'H-shift correlation beginning with the
'H and "C resonances of 6-H and C-6, respectively, which are shifted to
high field owing t o the benzyl resonance (S. Bradamante. G. A. Pagani, J.
Chem. SOC.Perkin Trans. 2 1986, 1035).
(1 e ' TMEDA), also has such short Li-H distances in the crystal (Li-H4
3.838, Li-HY 3.361, Li-HI I 3.437, Li-CH, 2.852, Li-CH, 3.240).
H.-J. Gais, J. Vollhardt, C. Kruger, Angew. Chem. 100 (1988) 1108; AngeM.
Chem. Int. Ed. Engl. 27 (1988) 1092.
Addition of Free Me,Ge: to Alkenes and Alkynes
on Glass Surfaces in the Presence of Water**
By Wilhelm P. Neumann,* Hideki Sakurai,* Gilbert Billeb,
Hartmut Brauer, Jiirgen Kocher, and Sabine Viebahn
Dedicated to Professor Christoph Riichardt on the occasion of
his 60fh birthday
nearly quantitatively3a; in the same way, reaction with D,O
and with H,180 gave [Dz]3a and [180]3a, respectively.
Indeed, water adsorbed at or in the glass[2' is involved
stoichiometrically in the reaction. This is evidenced additionally by cross checks: Using a quartz flask after heating it for
4 h at 150 "C/10-3 Torr, we found a yield of only 45% instead of 70%. Silylation of its inner surface by means of
F,C-CO-NMe-SiMe, (MSTFA)/Me,SiCI over 4 h at 70 "C
caused the yield of 3 a to decrease to 35%.
These surprising results required a reinvestigation of the
reactions of 1 with alkynes. We had obtained earlier with
phenylacetylene 4a a mixture of unsaturated four-, five-, and
six-membered germacycles.['] Erratically, a minor byproduct
with a I-alkenyl group had been observed ('H NMR), but
could not be identified at first. When we now added wet silica
gel before starting the reaction, high yields of the new digermoxane 5 a could be obtained instead of the: expected
products. Alkynes 4 b-d behaved analogously. The structures 5 have been clearly established, also by using H i 8 0 . It
should be noted that with 4c, for example, no product
could be detected earlier in the absence of H,O/SiO,.
Free Me,Ge: (dimethylgermanediyl or "dimethylgermylene") 1 undergoes a surprising variety of additions to IT systems of alkenes, 1,3-dienes, alkynes, and enones, most of
them being regio- or stereospecific.['] Now we have found
that reaction of 1 with conjugated substituted alkenes 2 or
with alkynes 4 gives products that contain two Ge atoms and
are generated by water adsorbed at the surface of glass or
silica gel.
2 Me,Ge
+ 2 HCEC-R
2 Me,Ge
+ 2 H,C=CH-X
(or D,O,
a, R = Ph; b, R = Bu; c, R = tBu; d, R
a, X = CN; b, X = COOMe; c ,
When we generated free 1 thermally in the presence of
acrylonitrile 2 a or other conjugated substituted olefins
(2 b, c) in a glass flask with the usual careful exclusion of air
and moisture, we obtained a moderate yield ofcompounds 3,
the hydrogen contents of which were higher than expected.
The additional hydrogens are not contributed via free radical
reactions between the partners or with the solvent, as suspected at first. This was established by the lack of influence of
cumene and triphenylmethane, as well as of tBuBr, which is
a powerful scavenger of Ge radicals. Finally, after several
difficulties, oxygen was found to be present in the product
and structure 3 was established. Compound 3, a dimer of the
adduct expected originally, contains an additional molecule
of water. Saturation of the solvent benzene with water did
not improve the yields of 3, and D,O gave no deuterated 3.
When we added finely powdered glass or silica gel saturated
with water and dried at
Torr, however, we obtained
Prof. Dr. W. P. Neumann, DiplLChem. G. Billeb, Dipl.-Chem. H. Brauer.
Dr. J. Kocher, Dip1:Chem. S. Viebahn
Lehrstuhl fur Orgdnische Chemie I der Universitat
D-4600 Dortmund 50 (FRG)
Prof. Dr. H. Sakurai
Department of Chemistry, Faculty of Science
Tohoku University, Sendai 980 (Japan)
This work was supported by the Fonds der Chemischen Industrie. We are
grateful to Dr. H. Hillgiirtner for very skilful1 mass spectrometric investigations.
VerlagsgesellschufrmhH, 0-6940 Weinheim, 1989
The germoxy group of 5 affords access to new organogermanium compounds, for example to the vinylgermyl acetate
6 starting from 5a and acetic anhydride. This is additional
evidence for the structure of compound 5 .
The structure of the product obtained from methyl acetylenecarboxylate 4 e and 1 was, at first, puzzling. The E-olefin
7 was found exclusively instead of the expected 1-alkenyl
compound corresponding to 5a-d.
2 Me,Ge
Elucidation of the mechanism of the participation of adsorbed water thus became a priority task. Compound 1
could react with water in a first step giving the previously
unknown hydride Me,Ge(H)OH, which could yield, via hydrogermylation, compounds 3, 5, and 7. In the absence of
alkene 2 or alkyne 4, however, water/silica gel and 1 gave
only the polygermane (Me,Ge),, which we futilely tried to
observe.[" Thus, the intermediacy of a germoxy compound
is excluded. An attempted in situ reaction of 1 with methanol
led to new 'H NMR absorptions of a hydride XMe,GeH,
but acrylonitrile 2 a, added consecutively, remained unchanged under the standard conditions. Further, no l-alkenyl structure corresponding to 5 could be observed when
phenylacetylene 4 a was hydrogermylated analogously by
Et,Ge(H)OMe; instead, the product of an inverse orientation was obtained.[31
Therefore, the first step is presumably a reaction between
1 and the alkene or alkyne. The resulting intermediate, which
Angew. Chem. lnt. Ed. Engl. 28 (1989) No. 8
Без категории
Размер файла
530 Кб
configurationally, lithiosulfones, stable
Пожаловаться на содержимое документа