Reactions of a silanediyl with carbon-oxygen and carbon-nitrogen double bonds

Article abstract of DOI:10.1021/jo951583b

Silanediyl 2 (generated by thermolysis of cyclotrisilane 1) reacts with benzophenone, tetracyclone, and fluorenone to yield products, which may originate from highly reactive siloxiranes as intermediates. However, using adamantanone as ketone, stable siloxirane 24 is obtained. The interaction of 2 with benzophenone anil or 36 gives heterocyclic compounds 31 and 37, respectively. The involvement of silaziridines in these reactions, as well as in the reactions of 1 with 1,4-diaza-1,3-butadienes 41a and b, which yield the expected formal [4 + 1] cycloaddition products, remains questionable.

Full text of DOI:10.1021/jo951583b

J . Org. Chem. 1996, 61, 3315-3319
3315
Rea ction s of a Sila n ed iyl w ith Ca r bon -Oxygen a n d
Ca r bon -Nitr ogen Dou ble Bon d s
J ohannes Belzner,* Heiko Ihmels, and Lara Pauletto
Institut f u¨ r Organische Chemie der Georg-August-Universit a¨ t G o¨ ttingen, Tammannstr. 2,
D-37077 G o¨ ttingen, Germany
Matthias Noltemeyer
Institut f u¨ r Anorganische Chemie der Georg-August-Universit a¨ t G o¨ ttingen, Tammannstr. 4,
D-37077 G o¨ ttingen, Germany
Received August 30, 1995X
Silanediyl 2 (generated by thermolysis of cyclotrisilane 1) reacts with benzophenone, tetracyclone,
and fluorenone to yield products, which may originate from highly reactive siloxiranes as
intermediates. However, using adamantanone as ketone, stable siloxirane 24 is obtained. The
interaction of 2 with benzophenone anil or 36 gives heterocyclic compounds 31 and 37, respectively.
The involvement of silaziridines in these reactions, as well as in the reactions of 1 with 1,4-diaza-
1
,3-butadienes 41a and b, which yield the expected formal [4 + 1] cycloaddition products, remains
questionable.
The intense study of the reactions of carbenes with
carbonyl compounds contrasts sharply with the quite
limited knowledge about the analogous reactions of
silanediyls, which is based mainly on the investigations
undertaken by Ando et al. In 1977, the first evidence
was presented for the formation of a transient siloxirane
the transfer of silanediyl 2 to imines and 1,4-diaza-1,3-
butadienes.
(
siloxacyclopropane) in the reaction of thermally gener-
1
ated dimethylsilanediyl with benzophenone. In 1982,
the same group succeeded in isolating and determining
the solid state structure of a stable, heavily substituted
siloxirane, which was obtained from the reaction of
dimesitylsilanediyl with 1,1,3,3-tetramethyl-2-indanone.2
Subsequent studies showed this strained three-mem-
bered ring to be built up via an initial electrophilic attack
of the silanediyl to the carbonyl oxygen, thereby forming
a silacarbonyl ylide as an intermediate.3 This reactive
intermediate undergoes ring closure to the corresponding
siloxirane, thus paralleling the mechanism of the forma-
tion of oxiranes in the reaction between carbenes and
ketones.4
Resu lts a n d Discu ssion
When 1 was stirred for 60 min at 60 °C with 3 equiv
of benzophenone, quantitative formation of siloxaindane
8
5
3
occurred; that is, silanediyl 2 shows the same chemi-
cal behavior as dimethylsilanediyl, which reacts with
benzophenone to yield an analogous siloxaindan. More-
We have reported earlier that cyclotrisilane 1 is in
equilibrium with presumably intramolecularly coordi-
nated silanediyl 2 at room temperature and acts toward
a variety of substrates as an effective synthetic equiva-
lent of silanediyl 2.5 Due to the mild conditions, under
which these silanediyl transfer reactions proceed, we
were able to synthesize stable strained ring systems such
1
over, the outcome of this reaction resembles that of the
addition of 2 to styrene, in which silaindan 4 was formed;
silacyclopropane 5 was detected and identified as an
intermediate of this reaction by means of NMR spectros-
copy.5 Thus, it is tempting to assume that the reaction
of 1 with benzophenone proceeds analogously forming a
siloxirane 8, which isomerizes via a vinylcyclopropane-
cyclopentene rearrangement and subsequent rearoma-
tization of 9 to yield heteroindan 3 (Scheme 1). However,
monitoring the reaction by NMR spectroscopy did not
provide any evidence for possible intermediates 8 or 9;
only signals that are attributable to starting material 1
and final product 3 were observed. An alternative
mechanistic pathway has to be taken into consider-
ation: Due to the possible intramolecular coordination
6
7
as silacyclopropenes, disilacyclobutenes, and silacyclo-
propanes.5 Thus, 1 appeared to be a promising precursor
for the synthesis of stable siloxiranes. Herein, we report
the results of reactions of 1 with ketones and will describe
X
Abstract published in Advance ACS Abstracts, April 15, 1996.
(
1) Ando, W.; Ikeno, M.; Sekiguchi, A. J . Am. Chem. Soc. 1977, 99,
447-6449.
2) Ando, W.; Hamada, Y.; Sekiguchi, A. Tetrahedron Lett. 1982,
3, 5323-5326.
3) Ando, W.; Hagiwara, K.; Sekiguchi, A. Organometallics 1987, 6,
270-2271.
4) Moss, R. A.; J ones, M. J r. In Reactive Intermediates; J ones, M.,
6
2
2
(
(
5
of silanediyl 2, it may act as a nucleophile as it is known
(
for tin(II) amides, which undergo nucleophilic attack to
J r., Moss, R. A., Eds.; Wiley: New York, 1985; Vol. 3, pp 45-108 and
related references cited therein.
acid chlorides.9
,10
Keeping in mind that the 2-((dimethy-
(
5) Belzner, J .; Ihmels, H.; Kneisel, B. O.; Gould, R. O.; Herbst-
lamino)methyl)phenyl substituent (Ar) shows a pro-
nounced tendency to coordinate to a positively charged
Irmer, R. Organometallics 1995, 14, 305-311.
(
(
6) Belzner, J .; Ihmels, H. Tetrahedron Lett. 1993, 34, 6541-6544.
7) Belzner, J .; Ihmels, H.; Kneisel, B. O.; Herbst-Irmer, R. J . Chem.
Soc., Chem. Commun. 1994, 1989-1990.
(8) Belzner, J . J . Organomet. Chem. 1992, 430, C51-C55.
S0022-3263(95)01583-0 CCC: $12.00 © 1996 American Chemical Society

3
316 J . Org. Chem., Vol. 61, No. 10, 1996
Belzner et al.
Sch em e 1
Sch em e 2
silicon center,11 one might speculate that the resulting
addition product 7 is stabilized by such a 2-fold intramo-
lecular coordination of its silicon terminus. Formation
of a stable Si-O bond and subsequent rearrangement of
the resulting siloxirane 8 via 9, as outlined in Scheme 1,
eventually opens the way from 7 to 3.
In order to suppress a [1,3] silicon shift in a possibly
formed siloxirane, tetracyclone was chosen as another
ketone: in this case, a vinylcyclopropane-cyclopentene
rearrangement of the resulting siloxirane 13 should not
be energetically feasible, because a strained bridgehead
olefin 15 would be generated (Scheme 2). The reaction
of 1 with 3 equiv of tetracyclone gave rise to the clean
formation of a single compound, but the appearance of a
to the carbonyl oxygen and subsequent ring closure of
the resulting sila ylide 10, or alternatively, via a nucleo-
philic addition of 2 to the carbonyl function of tetracy-
clone yielding 11, is not necessary in order to explain the
formation of 14: Its precursor 12 may be the product of
an electrocyclic ring closure of the initially generated sila
carbonyl ylide 10 as well.
1
singlet at δ 4.90 in the H NMR spectrum of the product
as well as a methine signal at δ 55.2 in the 13C NMR
spectrum was not in accordance with the expected
siloxirane structure 13. The complex NMR and mass
spectra did not allow an unambiguous identification of
the obtained product; however, X-ray diffraction analysis
revealed that compound 14 has been formed.12 Seem-
ingly, a siloxirane 13, if formed at all in this reaction,
now releases ring strain by undergoing a [1,7] silicon shift
to form 12, which yields 14 after a concluding [1,7]
hydrogen shift. Again, the involvement of a siloxirane
Fluorenone, too, turned out to be an inappropriate
starting ketone for the synthesis of a stable siloxirane
from 1: When 1 was reacted with 3 equiv of fluorenone,
a mixture of cyclotrisilane and the 2:1 adduct 16 was
obtained. The reaction was driven to completion by
addition of another 3 equiv of the ketone. The formation
of a 2:1 adduct from a ketone and a silanediyl is not
without precedents: Ando has reported that the pho-
tolytic generation of dimethylsilanediyl in the presence
1
3, which can be built up by an electrophilic attack of 2
(
9) Lappert, M. F.; Misra, M. C.; Onyszchuk, M.; Rowe, R. S.; Power,
P. P.; Slade, M. J . J . Organomet. Chem. 1987, 330, 31-46.
10) Kinetic investigations show that the reactions of 1 with
(
(12) Details of X-ray structure determination as well as tables of
atomic coordinates, bond lengths and angles, anisotropic displacement
parameters, and hydrogen atom coordinates of compounds 14 and 42a
have been deposited with the Cambridge Crystallographic Data Centre.
These data may be obtained, upon request, from the Director,
Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge,
CB2 1EZ, UK.
substituted styrenes and alkynes are favored by electron-withdrawing
substituents, i.e., the rate-determining step of the addition of silanediyl
2
to the multiple bond is of nucleophilic character: Belzner, J .; Ihmels,
H.; Dehnert, U. Unpublished results.
11) Belzner, J .; Sch a¨ r, D.; Kneisel, B. O.; Herbst-Irmer, R. Orga-
nometallics 1995, 14, 1840-1843.
(

Reactions of Silanedily with CdO and CdN Bonds
J . Org. Chem., Vol. 61, No. 10, 1996 3317
Sch em e 3
pathways to 21 have to be considered: 21 can be
generated directly by an electrophilic attack of 2 to the
carbonyl oxygen of fluorenone. Alternatively, the reac-
tion may be initiated by a nucleophilic attack of 2 to the
carbonyl center. Ring closure of 22 to form 23 and
subsequent heterolytic cleavage of a Si-C bond eventu-
ally results in the generation of 21; the driving force for
this reorganization would be the formation of a strong
Si-O bond as well as the aromatic stabilization of the
negative charge.15
Finally, the reaction of 1 with 3 equiv of adamantanone
allowed the synthesis of a stable compound, to which we
ascribe the structure of siloxirane 24. This result is
surprising in view of the fact that the reaction of the
bulkier dimesitylsilanediyl with the same ketone did not
stop at the stage of the corresponding siloxirane, but
instead yielded the 2:1 adduct 25.16 Especially valuable
for elucidating the cyclic structure of 24 are its NMR
data: The 29Si NMR signal is observed at δ -77.2 and
fits well into the shift range reported for other saturated
three-membered silacycles such as cyclotrisilanes or
silacyclopropanes.17 The H NMR as well as the C NMR
spectrum of 24 show, probably due to dynamic processes,
very broad signals at room temperature, which sharpen
up appreciably at elevated temperatures. The spiro
carbon atom absorbs at δ 84.6, which is in good ac-
cordance with δ 89.9 reported by Ando for a indanone-
1
13
of adamantanone results in the formation of two prod-
derived siloxirane.2
A H, H COSY spectrum of 24
1
1
ucts, 17 and 18.13 Again, a siloxirane 19, which is in
indicates that the geminal protons at C-3 and C-5 are,
presumably due to the anisotropic effect of the aryl
substituents at silicon, in quite different chemical envi-
ronments and thus give rise to an AB system, whereas
the protons located at C-7 and C-9, respectively, have
quite similar chemical shifts.
equilibrium with the ring-opened dipole 20 under the
reaction conditions, is assumed to be the crucial inter-
mediate. The orientation of the second adamantanone
unit in 17 is governed by the charge distribution of 20,
which is described by the resonance structures 20A and
Despite the chemical similarities between ketones and
imines, almost nothing is known about the reactions of
silanediyls with the latter compounds. Only in 1993,
Weidenbruch reported that photolytically generated di-
mesitylsilanediyl reacts with pyridine-substituted aldi-
mines 26 to yield a mixture of 27, which is formally the
insertion product of a silanediyl into the C-H bond of
2
0B. Similar arguments apply to the formation of 1,3-
dioxolanes as major products in the reaction of carbenes
with aldehydes.14 Thus, the exclusive formation of 16,
in which the incorporation of the second carbonyl unit is
exactly the opposite compared to 17, may be rationalized
best by the involvement of an ylide 21, in which negative
charge density is located at the carbon terminus whereas
the silicon center is positively charged. Indeed, it seems
reasonable to assume that a Lewis structure 21C con-
tributes the most to the ground state of 21 because it
benefits from a 2-fold stabilization: The negative charge
is delocalized into the fluorenyl substituent forming an
aromatic 14-electron system; moreover, the positively
charged silicon terminus of 21 can be stabilized by the
intramolecular coordination of the amino side arms of
both aromatic substituents. Again, two mechanistic
2
6, and 28, the formal [4 + 1] cycloaddition product; both
compounds are assumed to result from a rearrangement
18
of an initially formed silaziridine 29. In order to get
more insight into the reactivity of 2 toward imines, 1 was
stirred with 3 equiv of perphenylated ketimine 30, and
heteroindan 31 was obtained. The close structural
relationship between 8 and 31 suggests that 31 is formed
via a similar mechanistic pathway as 8 (cf. Scheme 1).
It is worth mentioning that the isomeric compound 32 is
not formed; 32 would result from a [1,3] silicon shift of
(
15) Another mechanistic pathway from 1 to 16 via radical inter-
mediates cannot be excluded, but seems less probable.
16) Ando, W.; Hamada, Y.; Sekiguchi, A. J . Chem. Soc., Chem.
Commun. 1983, 952-954.
17) See, e.g.: Williams, E. A. In The Chemistry of Organic Silicon
(
(
(
13) Ando, W.; Ikeno, M.; Sekiguchi, A. J . Am. Chem. Soc. 1978,
00, 3613-3615.
14) See, e.g.: de March, P.; Huisgen, R. J . Am. Chem. Soc. 1982,
04, 4592-4952.
Compounds; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, 1989;
pp 511-554.
(18) Weidenbruch, M.; Piel, H.; Peters, K.; von Schnering, H. G.
Organometallics 1993, 12, 2881-2882.
1
1
(

3
318 J . Org. Chem., Vol. 61, No. 10, 1996
Belzner et al.
When the C-N double bond of the starting imine
compound is part of a conjugated system as in 1,4-diaza-
1
,3-butadienes 41a and b, the corresponding diazasila-
1
9
cyclopentenes 42a and b were obtained quantitatively,
as one would expect in view of the fact that di-tert-
butylsilanediyl underwent the analogous reaction with
4
1a .20 These five-membered rings, which are the formal
product of a [4 + 1] cycloaddition reaction between
silanediyl and heterodiene, may result from a [1,3] silicon
shift of a primarily formed vinylsilaziridine, thus resem-
bling the synthesis of silacylo-3-pentenes by reaction of
silanediyls with conjugated dienes, which is known to
2
1
proceed via intermediate vinylsilacyclopropanes. How-
ever, no spectroscopic evidence for the initial formation
of vinylsilaziridine in the reaction with azabutadienes
could be obtained.
an intermediate silaziridine 33 forming a new bond to
the ortho carbon of the phenyl substituent, which is
located at the nitrogen, and a subsequent [1,3] hydrogen
shift. The exclusive formation of 31 may be due to the
fact that this [1,3] silicon shift requires the breaking of
a Si-N bond in 33; thus, an alternative [1,3] silicon shift,
which proceeds by cleavage of the less stable Si-C bond
and finally results via 34 in the formation of 31, seems
to be energetically more favorable. Alternatively, one
might speculate that silaziridine 33 is not the precursor
of 34, but rather the initially formed sila ylide 35
undergoes direct ring closure to 34.
In conclusion, a variety of silicon-containing hetero-
cycles was synthesized by interaction of cyclotrisilane 1
with ketones and imines. The possibility of the transient
existence of a siloxirane during the reaction of 1 with
benzophenone, tetracyclone, or fluorenone is corroborated
by the successful isolation of a siloxirane 24, whereas the
search for a stable silaziridine has still to go on. The
question whether the attack of silanediyl 2 to C-O and
C-N double bonds is of electrophilic or nucleophilic
nature remains a matter, which is currently being
adressed by our group.
Exp er im en ta l Section
Gen er a l Meth od s. 1H-NMR and C-NMR spectra were
13
1
13
29
recorded at 250 MHz ( H), 62.9 MHz ( C), and 59.6 MHz ( Si).
Mass spectra were recorded using electron impact (EI, 70 eV)
and fast atom bombardment (FAB) mode; HRMS were deter-
mined using preselected ion peak matching at R ∼ 10 000 to
be within (2 ppm of the exact mass. Melting points are
uncorrected. Elemental analyses were performed at Mi-
kroanalytisches Labor der Georg-August-Universit a¨ t G o¨ ttin-
gen.
Contrasting the structural similarity between reaction
products 8 and 31, the reaction of 2 with fluorene-derived
imine 36 took another course than with fluorenone:
Instead of a 2:1 adduct between imine and silanediyl,
corresponding to the formation of 16, the 1:1 adduct 37
was obtained. Presumably, the steric strain imposed to
3
6 due to the bulky 2,6-dimethylphenyl substituent does
All manipulations were carried out under an inert argon
atmosphere using carefully dried glassware. Solvents used
were dried by refluxing over sodium and distilled immediately
before use.
not allow the attack of sila ylide 38 to a second molecule
of 36. Thus, 38 or silaziridine 39 undergoes rearrange-
ment to 40; a concluding [1,3] silicon shift eventually
restores the aromatic fluorene system and yields 40.
8
Hexakis[2-((dimethylamino)methyl)phenyl]cyclotrisilane (1),
benzophenone anil (30),22 fluorenone 2′,6′-dimethylanil (36),
1
22
2
3
,4-di-tert-butyl-1,4-diaza-1,3-butadiene (41a ), and 1,4-dicy-
clohexyl-1,4-diaza-1,3-butadiene (41b) were prepared accord-
2
3
ing to published procedures.
1
,1-Bis[2-((d im eth yla m in o)m eth yl)p h en yl]-3-p h en yl-2-
oxa -1-sila in d a n (3). A solution of 123 mg (0.14 mmol) of 1
and 78 mg (0.43 mmol) of benzophenone in 10 mL of toluene
was heated for 60 min at 50 °C. The solvent was removed in
vacuo, and the residue was crystallized from 10 mL of
n-hexane at -6 °C to give 109 mg (55%) of 3 as colorless
1
crystals: mp 115-117 °C; H NMR (C
6
D
6
) δ 1.80 (s, 6 H), 1.85
(19) The solid state structure of 42a (determined by X-ray crystal-
lography) does not differ significantly from that reported by Weiden-
2
0
bruch for another diazasilacyclopentene.
20) Weidenbruch, M.; Lesch, A.; Peters, K. J . Organomet. Chem.
(
1
991, 407, 31-40.
(
(
(
21) See ref 5 and references cited therein.
22) Reddelien, G. Chem. Ber. 1913, 46, 2718-2723.
23) Kliegman, J . M.; Barnes, R. K. Tetrahedron 1970, 26, 2555-
2
560.

Reactions of Silanedily with CdO and CdN Bonds
J . Org. Chem., Vol. 61, No. 10, 1996 3319
(
s, 6 H), 2.92, 3.61 (2 × d, J ) 13 Hz, 2 H), 3.23, 3.24 (2 × d,
129.2, 129.8, 130.3, 130.5, 132.4, 132.7, 134.0, 137.7, 138.5,
2
9
J ) 13 Hz, 2 H), 6.36 (s, 1 H), 6.94-7.33 (m, 14 H), 7.96-8.02
(
6
1
134.0, 144.2, 145.8, 146.9, 152.0; Si NMR (C
6
D
6
) δ -2.8 (br);
1
3
+
+
m, 2 H), 8.09-8.18 (m, 1 H); C NMR (CDCl
3
) δ 45.3, 45.4,
4.2, 64.7, 83.2, 124.2, 126.8, 126.8, 126.9, 127.5, 127.7, 128.3,
28.5, 128.9, 129.1, 129.2, 129.9, 133.8, 135.4, 135.7, 137.5,
MS (EI) m/ z 553 (21) [M ], 419 (52) [M - Ar], 297 (31) [Ar -
2
+
+
Si + H], 58 (100) [CH
2
NMe
2
]; HRMS calcd for C37
H
39
N
3
Si
553.2913, found 553.2913. Anal. Calcd for C37
80.24; H, 7.10. Found: C, 80.08; H, 7.20.
39 3
H N Si: C,
2
9
1
-
38.1, 137.8, 144.6, 145.5, 146.3, 153.7; Si NMR (C
6
D
6
) δ
9.20; MS (EI) m/ z 478 (3) [M ], 344 (100) [M - Ar]; HRMS
OSi 478.2440, found 478.2440.
,5-Bis[2-((d im et h yla m in o)m et h yl)p h en yl]-1,2,3-t r i-
+
+
2,2-Bis[2-((d im et h yla m in o)m et h yl)p h en yl]-1-(2,6-d i-
m eth ylp h en yl)-1-a za -2-sila -1,2-d ih yd r oa ceflu or en ylen e
(37). A solution of 233 mg (0.26 mmol) of 1 and 221 mg (0.78
mmol) fluorenone 2,6-dimethylanil (36) in 20 mL of toluene
was stirred for 12 h at 50 °C. Toluene was removed in vacuo,
and the residue was crystallized from n-hexane to give 349
calcd for C31
34 2
H N
5
p h en yl-1,5-d ih yd r o-4-oxa -5-sila cyclop en t a [a ]n a p h t h a -
lin e (14). To a solution of 401 mg (0.45 mmol) of 1 in 10 mL
of toluene was added 529 mg (1.38 mmol) of tetracyclone. After
the solution was stirred for 14 h at rt, toluene was replaced
1
mg (77%) of 37 as a white solid: mp 164 °C; H NMR (CDCl )
3
by 8 mL of Et
remaining insoluble violet solid was crystallized from THF/
Et O (1:1) to yield 614 mg (67%) of 14 as pale yellow crystals:
mp 168-170 °C; H NMR (CDCl
H), 3.26, 3.32 (2 × d, J ) 14 Hz, 2 H), 3.39 (d, J ) 14 Hz, 2 H),
2
O. The resulting slurry was filtered, and the
δ 1.50 (s, 3 H), 1.63 (s, 12 H), 1.91 (s, 6 H), 2.26 (s, 3 H), 2.03,
3.02 (2 × d, J ) 15 Hz, 2 H), 2.59, 3.21 (2 × d, J ) 13 Hz, 2
H), 5.90 (s, 1 H), 6.68 (d, J ) 8 Hz, 1 H), 6.78 (dd, J ) 6, 3 Hz,
1 H), 6.92-7.02 (m, 4 H), 7.21-7.46 (m, 8 H), 7.59 (d, J ) 7
Hz, 1 H), 7.67 (d, J ) 7 Hz, 2 H), 8.19 (dd, J ) 7, 2 Hz, 1 H);
2
1
3
) δ 1.66 (s, 6 H), 1.81 (s, 6
1
3
3
.58 (d, J ) 14 Hz, 2 H), 4.90 (s, 1 H), 6.88-7.53 (m, 27 H);
C NMR (CDCl ) δ 20.3, 20.5, 44.9, 45.3, 62.2, 64.3, 70.3, 121.1,
3
1
3
C NMR (CDCl
3
) δ 44.8, 45.1, 55.2, 63.8, 63.8, 121.3-154.8;
121.2, 124.9, 125.2, 125.4, 126.2, 126.4, 126.8, 127.4, 128.2,
128.5, 128.7, 129.0, 129.7, 131.6, 133.7, 134.4, 135.6, 136.7,
136.8, 136.8, 138.0, 140.2, 141.9, 144.4, 145.3, 145.6, 146.6,
2
9
+
Si NMR (CDCl ) δ -13.4; MS (EI) m/ z 680 (16) [M ], 635
3
+
+
+
(
7) [M - HNMe ], 546 (100), [M - Ar], 412 (16) [M - 2
2
2
+
+
+
29
Ar], 340 (7) [M ], 134 (4) [Ar ], 58 (15) [CH
Calcd for C47 OSi: C, 82.90; H, 6.51; N, 4.11. Found: C,
2.78; H, 6.58; N, 3.83.
,1-Bis[2-((d im eth yla m in o)m eth yl)p h en yl]-3,4-bis(flu -
2
NMe
2
]. Anal.
148.6, 162.6; Si NMR (CDCl ) δ -0.2; MS (EI) m/ z 579 (10)
3
+
+
+
H
44
N
2
[M ], 534 (100) [M - HNMe ], 489 (71) [M - 2 HNMe ], 476
(52) [M - HNMe - CH NMe ]; HRMS calcd for C H N Si
2
2
+
8
2
2
2
39 41
3
1
579.3070, found 579.3069.
,1-Bis[2-((d im eth yla m in o)m eth yl)p h en yl]-2,5-d i-ter t-
bu tyl-2,5-d ia za -1-sila cyclop en t-2-en e (42a ). A solution of
.497 g (1.68 mmol) of 1 and 850 mg (5.05 mmol) of 41a in 20
or en ylid en e)-2,5-d ioxa -1-sila cyclop en ta n e (16). A solu-
tion of 169 mg (0.19 mmol) of 1 and 205 mg (1.14 mmol) of
fluorenone in 20 mL of toluene was stirred for 4 h at 50 °C.
After concentration of the solution in vacuo, a yellow solid
crystallized at 0 °C, which was washed with 5 mL of n-hexane.
The solid was dried in vacuo to give 256 mg (68%) of 16 as
yellow solid: mp >200 °C; H NMR (CDCl
3
1
1
mL of toluene was stirred for 3 h at 60 °C. After removal of
1
toluene in vacuo, H NMR spectroscopically pure 42a remained
as a green solid, which was crystallized from toluene to give
1
3
) δ 1.95 (s, 12 H),
1
1
.243 mg (53%) of 42a as analytically pure green crystals: mp
.41 (s, 4 H), 6.72 (dd, J ) 7 Hz, 7 Hz, 4 H), 6.92-7.10 (m, 12
1
75 °C; H NMR (CDCl ) δ 1.01 (s, 18 H), 2.32 (s, 12 H), 3.82
3
H), 7.35-7.39 (m, 2 H), 7.53 (mc, 4 H), 8.63-8.68 (m, 2 H);
(s, 4 H), 5.88 (s, 2 H), 7.17 (dd, J ) 7, 7 Hz, 2 H), 7.42 (dd, J
) 8, 7 Hz, 2 H), 7.83 (d, J ) 8, 2 H), 7.88 (d, J ) 7 Hz, 2 H);
13
1
3
C NMR (CDCl
3
) δ 45.7, 63.8, 93.5, 118.8 (br), 125.2 (br), 126.4,
29
1
27.1, 127.6, 130.1, 137.1, 138.0, 140 (very br), 145.4; Si NMR
C NMR (CDCl
3
) δ 30.2, 46.0, 51.8, 62.6, 112.2, 124.0, 127.3,
29
+
+
(
CDCl
HNMe
HRMS)], 180 (100) [fluorenone ], 58 (19) [CH
calcd for C44 Si 656.2859, found 656.2859.
′,2′-Bis[2-((d im et h yla m in o)m et h yl)p h en yl]-3′-oxa -2′-
3
) δ -31.5; MS (EI) m/ z 656 (1) [M ], 611 (2) [M
], 522 (1) [M - Ar], 476 (78) [M - fluorenone
-
1
29.8, 133.5, 137.6, 148.8; Si NMR (CDCl ) δ -22.6; MS (EI)
3
+
+
+
+
+
2
m/ z 464 (100) [M ], 407 (2) [M - t-Bu], 330 (7) [M - Ar].
Anal. Calcd for C27 Si: C, 72.36; H, 9.54; N, 12.05.
Found: C, 72.47; H, 9.64; N, 12.14.
,1-Bis[2-((d im eth yla m in o)m eth yl)p h en yl]-2,5-d icyclo-
h exyl-2,5-d ia za -1-sila cyclop en t-2-en e (42b). A solution of
8 mg (0.08 mmol) of 1 and 50 mg (0.23 mmol) of 41b in 0.4
+
+
2 2
NMe ]; HRMS
(
44 4
H N
40 2 2
H N O
2
1
sila a d a m a n ta n esp ir ocyclop r op a n e (24). A solution of 255
mg (0.29 mmol) of 1 and 129 mg (0.86 mmol) of adamantanone
in 10 mL of toluene was stirred for 12 h at 60 °C. Toluene
was removed in vacuo, and 10 mL of Et O was added to the
2
residue to give a white slurry. After filtration 179 mg of 24
was obtained as a white solid. Another 123 mg (total 72%) of
2
(
d, J ) 11 Hz, 4 H, 3-H, 5-H), 1.90 (br s, 1 H), 1.91-2.02 (m,
2
8
2
6
1
mL of C
6 6
D was heated for 12 h at 50 °C to give H NMR
spectroscopically pure 42b as viscous green oil; attempts to
1
crystallize it using a variety of solvents were unsuccessful:
NMR (C
H
6
D
6
) δ 0.83-1.91 (m, 20 H), 2.03 (s, 12 H), 3.26-3.49
1
4 crystallized from the filtrate: mp 150-152 °C; H NMR
(
(
m, 2 H), 3.51 (s, 4 H), 5.99 (s, 2 H), 7.10-7.27 (m, 4 H), 7.60
1
1
323 K, C
6
D
6, additional H, H COSY, 333 K) δ 1.67, 2.84 (2 ×
13
d, J ) 7 Hz, 2 H), 7.96 (d, J ) 7 Hz, 2 H); C NMR (C
6
D
6
) δ
2
1
5
6.2, 26.7, 34.4, 45.6, 53.3 , 64.2, 112.3 , 126.5, 129.4, 129.8,
0 H), 3.07, 3.52 (2 × d, J ) 13 Hz, 4 H), 7.16-7.21 (m, 6 H),
29
36.1, 136.7, 145.7; Si NMR (C
6
D
6
) δ -17.6; MS (EI) m/ z
.15-8.17 (m, 2 H); 13C NMR (328 K, 75.5 MHz, C
D
) δ 28.4,
6
6
+
+
16 (100) [M ], 382 (10) [M - Ar]; HRMS calcd for C32H N -
48 4
9.1, 33.7, 38.1, 38.6, 45.8, 64.3, 84.6 (C-1), 126.4, 129.2 , 135.5,
36.5, 144.7, one ar-CH was not detected, probably due to
Si 516.3648, found 516.3648.
1
2
9
overlap with C
3
4
6
D
6
-signals; Si NMR (C
6
D
6
) δ -77.2; MS (FAB,
+
Ack n ow led gm en t. This work was supported by the
Deutsche Forschungsgemeinschaft (financial support,
fellowship to J .B.), the Fonds der Chemischen Industrie
(financial support), and the Friedrich-Ebert-Stiftung
(fellowship to H.I.). L.P. was supported by an ERAS-
MUS exchange grant from the European Community.
-NBA) m/ z (relative intensity) 600 (30) [M + matrix + H],
65 (82) [M + matrix - Ar], 447 (100) [M + H], 402 (25)
+
+
+
[
M - NMe
N, 6.27. Found: C, 75.13; H, 8.59; N, 6.33.
,5-Ben zo-1,1-bis[2-((d im eth yla m in o)m eth yl)p h en yl]-
,3-d ip h en yl-2-a za -1-sila cyclop en t-4-en e (31). A solution
2 38 2
]. Anal. Calcd for C28H N OSi: C, 75.29; H, 8.57;
4
2
of 127 mg (0.14 mmol) of 1 and 108 mg (0.42 mmol) of
benzophenone anil was stirred for 12 h at 50 °C. The solvent
was removed in vacuo, and the residue was washed with 10
mL of Et
1
2
1
Su p p or tin g In for m a tion Ava ila ble: Copies of H spectra
of compounds for which no elemental analysis was obtained,
2
O to yield 100 mg (43%) of 31 as a white solid: mp
1
1
a copy of a H, H COSY spectrum of 24, and a listing of NMR
data with subjective peak assignments (9 pages). This mate-
rial is contained in libraries on microfiche, immediately follows
this article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
1
82-184 °C; H NMR (CDCl
3
) δ 1.51 (s, 6 H), 2.14 (s, 6 H),
.64, 3.06 (2 × d, J ) 13 Hz, 2 H), 3.50, 3.79 (2 × d, J ) 14
Hz, 2 H), 5.85 (s, 1 H), 6.48 (dd, J ) 7, 7 Hz, 1 H), 6.79-6.89
(
m, 4 H), 7.00-7.25 (m, 10 H), 7.37-7.43 (m, 3 H), 7.59-7.74
13
(m, 3 H), 8.00 (d, J ) 7 Hz, 1 H); C NMR (CDCl
3
) δ 45.2,
5.4, 63.5, 65.5, 68.9, 118.2 (3 signals), 125.5, 126.1, 126.3,
26.6, 127.0, 127.9, 128.2, 128.5 (2 signals), 128.5 (2 signals),
4
1
J O951583B