Small rings, 93 - 1,2,3,4-Tetra-tert-butyl-4-trimethylsilyl-4-sila-2- cyclobuten-1-yl: A quantum mechanical and ESR study

Article abstract of DOI:10.1002/1099-0690(200004)2000:7<1107::AID-EJOC1107>3.0.CO;2-8

Trisilane 7 has been synthesized as a potential photochemical precursor of tetra-tert-butylsilatetrahedrane (5) and/or tetra-tert- butylsilacyclobutadiene (6). Astonishingly, only one Si,Si bond is broken upon irradiation of 7 and the silacyclobutenyl radical 9 can be identified as the reaction product. The structure of radical 9, which in the absence of oxygen is persistent even at room temperature, has been elucidated by comparison of its experimental and calculated ESR spectra.

Full text of DOI:10.1002/1099-0690(200004)2000:7<1107::AID-EJOC1107>3.0.CO;2-8

FULL PAPER
Small Rings, 93[‡]
1,2,3,4-Tetra-tert-butyl-4-trimethylsilyl-4-sila-2-cyclobuten-1-yl:
A Quantum Mechanical and ESR Study
Günther Maier,*[a] Armin Kratt,[a] Andreas Schick,[a] Hans Peter Reisenauer,[a]
[b]
[b]
´ ´
Frederique Barbosa, and Georg Gescheidt*
Keywords: Photochemistry / Radicals / ESR spectroscopy / Quantum mechanical calculations / Silicon compounds
Trisilane 7 has been synthesized as a potential photochem- clobutenyl radical 9 can be identified as the reaction product.
ical precursor of tetra-tert-butylsilatetrahedrane (5) and/or
tetra-tert-butylsilacyclobutadiene (6). Astonishingly, only
The structure of radical 9, which in the absence of oxygen is
persistent even at room temperature, has been elucidated by
one Si,Si bond is broken upon irradiation of 7 and the silacy- comparison of its experimental and calculated ESR spectra.
Introduction
In a previous report on the synthesis and properties of
tetra-tert-butyltetrahedrane (3) and tetra-tert-butylcyclobu-
tadiene (4),[1] the corresponding sila derivatives 5 and 6
were envisaged as worthwhile goals for forthcoming studies.
The methodology for their preparation stems from experi-
ence gained in synthesizing 3 and 4, which was achieved
starting from the cyclopropenyldiazomethane 1.[2] Photo-
chemical nitrogen elimination from 1 leads to carbene 2.
This intermediate cannot be detected since it undergoes
rapid ring-expansion to cyclobutadiene 4 or is directly con-
verted to tetrahedrane 3 through a cheletropic cycloaddi-
tion of the carbenic center to the double bond of the three-
membered ring.[2] Another route to 3 involves cross-addi-
tion of the double bonds in 4 upon photochemical excita-
tion.[1]
This cyclopropenylcarbinyl route, which has hitherto
been successfully exploited in the synthesis of several other
alkyl-,[3] silyl-,[4] and germyl-substituted[5] analogs of 3 and
4, was also envisaged as being feasible with one carbon in
the ring skeleton replaced by a silicon atom. Thus, we initi-
Neither silatetrahedrane nor silacyclobutadiene could be
ated a project directed towards the preparation of 5 and 6.[6]
isolated in a matrix at 10 K.[8] This encouraged us to look
According to calculations (B3LYP/6-31G*) on the
for the kinetically stabilized derivatives 5 and 6.[10]
C3H4Si energy hypersurface,[7,8] the parent silacyclobutadi-
ene, cyclopropenylsilylene, and silatetrahedrane are 47.8,
Synthesis of the Persistent Radical 9
54.7, and 71.2 kcal mol–1, respectively, higher in energy
than the global minimum, 2-methyl-1-silacyclopropenylid-
ene. This order of relative energies is in agreement with cal-
culations carried out previously by Gordon et al.[9]
In the silicon series, it is not possible to start from a diazo
compound akin to 1. If one wants to generate silylene 10,
it is adequate to start with trisilane 7.[6] Irradiation of 7 in
cyclohexane at room temperature at a wavelength of 254 nm
generates a yellow solution showing an absorption max-
imum at λmax ϭ 476 nm. At first, we believed that the com-
pound responsible for the yellow color might be silylene 10.
However, it turned out that the absorption maximum is in
fact due to silacyclobutenyl radical 9. The structure of 9
was established on the basis of its ESR spectrum (see be-
low) and the results of trapping experiments. Radical 9 is
[‡]Part 92: G. Maier, A. Bothur, Eur. J. Org. Chem. 1998,
2063–2072.
[a]
Institut für Organische Chemie der Universität Giessen,
Heinrich-Buff-Ring 58, D-35392 Giessen, Germany
Fax: (internat.) ϩ49 (0)641/99-34309
[b]
Department Chemie, Universität Basel,
Klingenbergstraße 80, CH-4056 Basel, Switzerland
Fax: (internat.) ϩ41 (0)61/2673855
E-mail: georg.gescheidt@unibas.ch
Eur. J. Org. Chem. 2000, 1107Ϫ1112
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ193X/00/0407Ϫ1107 $ 17.50ϩ.50/0
1107

G. Maier, A. Kratt, A. Schick, H. P. Reisenauer, F. Barbosa, G. Gescheidt
FULL PAPER
persistent provided that no oxygen is present. Even after geometry was optimized using the semiempirical AM1[12]
two years the ESR spectrum of the yellow solution of 9 and ab initio UHF/3-21G* methods without any symmetry
could still be measured.
restrictions. Both calculations led to a comparable geo-
metry, in good accord with the Cs symmetry of 9. Averaged
values of bond lengths, bond angles, and dihedral angles
are compared in Figure 1. In view of the good agreement,
we will limit the description of the geometry of 9 to the
data obtained using the UHF/3-21G* procedure. The C(1)–
˚
C(2) and C(2)–C(3) distances of 1.456 A are typical for a
delocalized π system. All carbon–silicon bond lengths are
similar and are analogous to those calculated for the closed-
[13]
˚
˚
shell silacyclobutane (1.931 A).
The same holds for the
Si(4)–Si(5) distance of 2.387 A. Typical Si–Si bond lengths
[14]
˚
lie between 2.33 and 2.39 A.
The C(1)–C(2)–C(3) bond
angle of ca. 103° illustrates the bending of the three allylic
π centres. As a consequence, the C(1)–Si(4)–C(3) angle at
the Si corner of the silacyclobutane measures 74.5°. The
central silacyclobutane moiety is only marginally displaced
from a planar arrangement. The Si(4) atom is bent out of
the C(1)–C(2)–C(3) plane [towards C(6)] by 8.1°.
Exposure to oxygen results in immediate quenching of
the yellow color of a solution of 9 and lactone 13 can be
isolated as the reaction product. A much better structural
proof than the transformation of 9 into 13 is the more
straightforward reaction of 9 with ethanethiol. This reagent
transfers a hydrogen atom to 9 and one diastereomer of
cyclobutene 12 can be isolated.
How is radical 9 formed? Clearly, irradiation of 7 does
not initiate an extrusion yielding silylene 10 and finally 5 or
6.[10] Instead, a homolytic cleavage of the Si–SiMe2Ph bond
occurs.[11] The initially formed radical 8 then rearranges to
the more stable cyclobutenyl radical 9.
Direct experimental proof may be obtained for the inter-
mediacy of monoradical 8. Thus, when the irradiation of
trisilane 7 is carried out on a preparative scale, the main
reaction product is disilane 11, which can be isolated in 62%
yield together with 9% of lactone 13. The cyclopropyl deriv-
ative 11 has to be regarded as the product of a combination
of 8 with a hydrogen atom, which is split off from the solv-
ent or from a tert-butyl group of one of the dissolved mole-
cules.
Figure 1. Geometrical parameters of 9
The SOMO of 9 closely resembles that of the allyl rad-
ical. Based on the above geometries, the hyperfine coupling
constants (hfc) were determined by single-point density
functional theory calculations[15–18] at the unrestricted level
of theory using the B3LYP functional[19] (Table 1).
ESR Spectrum of Radical 9
The observation of 9 is consistent with the results of Fink
et al.,[10c] who detected triplet diradicals similar to mono-
radical 9 as intermediates during the dimerization of aryl
tri-tert-butylsilacyclobutadienes.
A very good agreement between the experimental and
simulated ESR spectra was achieved when the ESR signals
were attributed to a superposition of five spectral compon-
ents (Figure 2). The most intense component is a central
19-line pattern indicative of an interaction between the un-
Structure of Radical 9
1
paired electron and 18 equivalent protons with a H hyper-
In order to gain further insight into the structure of rad- fine coupling constant of 0.037 mT. Thus, this coupling
ical 9, we performed quantum mechanical calculations. The constant can be assigned to the 18 equivalent γ protons of
1108
Eur. J. Org. Chem. 2000, 1107Ϫ1112

1,2,3,4-Tetra-tert-butyl-4-trimethylsilyl-4-sila-2-cyclobuten-1-yl
the tert-butyl groups at C(1) and C(3). This assignment is
FULL PAPER
The inset in Figure 2 shows that an additional low-intens-
in very good accord with the calculated values (Table 1). ity ESR line may be present between the two outermost
The outermost components have the same line pattern as satellite lines. This would be consistent with the 13C hfc cal-
the central signal with a relative intensity of ca. 2.2%. This culated for C(2) (Table 1). However, because of the low in-
corresponds to hyperfine coupling with a 13C nucleus (ms ϭ tensity of this line (one 13C nucleus, i.e. 1.1%), an unam-
1/2; hfc ϭ 2.560 mT) present in two equivalent positions of biguous simulation is not possible.
9 (the natural abundance of 13C is 1.1%). The simulation
The hyperfine data derived from the experimental ESR
for the next doublet, which is spaced by 1.495 mT and is spectrum are in satisfying accord with the calculated data
1
ca. twice as intense as the outermost lines, reveals a 4.7% (Table 1). The assignment of the H hfc of 0.037 mT to the
intensity relative to the central feature and matches the nat- 18 equivalent Hγ is as straightforward as that of the prom-
ural abundance of the 29Si nucleus (4.7%). The main prob- inent 13C hfc of 2.56 mT to the equivalent C(1) and C(3)
1
lem was the analysis of the doublet of satellite lines seen nuclei of the allylic moiety. A similar H hfc of 0.027 mT
closest to the central multiplet. Two possibilities were con- has previously been found for the radical cation of cyclobu-
sidered: (i) coupling with one type of nucleus (either 29Si or tadiene 4.[20] The predicted 29Si hfc value for Si(4) in 9 of
13C) or (ii) interaction with two nuclei, 29Si and 13C, leading –1.299 mT (–1.194 mT; here, the values based on the ab in-
to overlapping satellite lines. No satisfactory agreement itio geometry given in the text are followed by those based
could be obtained when only one 29Si [Si(5)] nucleus or six on AM1 geometry in parentheses) is closely matched by the
equivalent 13C nuclei (Cγ) were taken into account. The corresponding experimental value of 1.495 mT.
best fit was achieved for two components with rather sim-
Even the close similarity between the 13C hfcs of six and
ilar 13C hfcs: one with a 13C hfc of 0.692 mT stemming two equivalent nuclei is convincingly accounted for by the
from the interaction with six equivalent nuclei (6.6%) and calculations. The 13C hfc of the two equivalent Cβ nuclei is
one with a 13C hfc of 0.656 mT (2.2%). Although the 3:1 predicted to be –0.656 mT (–0.630 mT) and that of the six
intensity ratio of these two components has to be utilized equivalent Cγ ϩ0.661 mT (ϩ0.798 mT); this is mirrored by
for a successful simulation, a 5% shift in the 13C hfcs does experimental values of 0.656 and 0.692 mT, respectively.
not significantly alter the correlation between the experi- The opposite signs of the latter two 13C hfcs imply two dif-
mental and simulated spectra.
ferent paradigms of spin delocalization. Whereas the spin
transfer to Cβ proceeds through spin polarization,[21] thus
leading to an excess of β spin (negative coupling constant),
the spin population detected at Cγ is transferred by C,C
hyperconjugation inducing an excess of α spin.
Application of the ENDOR technique[22] revealed a pair
1
of broad peaks corresponding to the H hfc of 0.037 mT
1
and presumably some smaller H hfcs not resolved in the
ESR signal. Attempts to determine the hfcs stemming from
the 13C and 29Si nuclei by ENDOR were unsuccessful. All
the remaining calculated hfcs irrelevant for a clear-cut iden-
tification are included in Table 1.
Synthesis of Precursor 7
Starting from tert-butylphenylsilane 14,[23,24] precursor 7
was synthesized as outlined in Scheme 1. Chlorination of
14 was performed using PCl5 in CCl4.[24] Treatment of chlo-
rosilane 15 with lithium in THF and subsequent addition
of trimethylchlorosilane to the resulting silyl anion yielded
disilane 16.[24] In order to remove the phenyl group, 16 was
stirred with liquid HBr at –78 °C. Trisilane 18 could be
obtained by reaction of 17 with the dimethylphenylsilyl an-
ion, which was prepared from chlorodimethylphenylsilane
and lithium. Transformation of 18 into 19 was readily
achieved by stirring in CCl4 at 50 °C for 20 h. Finally, the
cyclopropenyl substituent was introduced by treating
chlorotrisilane 19 with lithium and adding the thus gener-
ated anion solution to a suspension of tri-tert-butylcyclo-
propenylium tetrafluoroborate in THF in a dropwise fash-
ion. The product was purified by preparative TLC and
recrystallization.
Figure 2. Experimental and simulated (WINSIM) ESR spectra of
9; the inset shows the satellite lines in the flanks of the main central
signal on an expanded scale; the numbers below the simulated spec-
trum indicate the relative percentages of satellite lines with respect
to the whole integrated spectrum; the arrow in the inset indicates
the position at which a signal attributable to hyperfine coupling
with C(2) could be present
Eur. J. Org. Chem. 2000, 1107Ϫ1112
1109

G. Maier, A. Kratt, A. Schick, H. P. Reisenauer, F. Barbosa, G. Gescheidt
Table 1. Calculated and experimental hyperfine coupling constants (hfcs) attributed to 9
FULL PAPER
Nucleus
Calcd. hfcs [mT]
UB3LYP/6-31G*//UHF/3-21G*
Calcd. hfcs [mT]
UB3LYP/6-31G*//AM1
Experimental
hfcs [mT][a]
C(1), C(3)
ϩ3.258
Ϫ1.299
ϩ0.661[b]
Ϫ0.656
ϩ0.038[b]
Ϫ1.851
ϩ0.154
ϩ0.099
Ϫ0.164
ϩ0.003
ϩ0.009
Ϫ0.026
ϩ0.004[b]
ϩ0.014[b]
ϩ0.002[b]
ϩ2.961
Ϫ1.194
ϩ0.798[b]
Ϫ0.630
ϩ0.047[b]
Ϫ1.601
ϩ0.134
ϩ0.048
Ϫ0.156
Ϫ0.006
ϩ0.015
Ϫ0.022
Ϫ0.007[b]
ϩ0.001[b]
ϩ0.004[b]
2.560
1.495
0.692
0.656
0.037
Si(4)
Cγ
Cβ
Hγ
C(2)
CβЈ
C–Si(4)
CγЈ
(CH3)3C–Si(4)
(CH3)3Si–Si(4)
Si(5)
HγЈ
(CH3)3C–Si(4)
(CH3)3Si–Si(4)
[a]
[b]
Determined by simulation (WINSIM). – Values averaged over all rotationally different protons or 13C nuclei.
Experimental Section
General: All reactions were carried out under dry argon. Solvents
were dried using standard techniques. All glassware was thoroughly
oven-dried at 130 °C prior to use. – NMR: Bruker AM 400 or AC
200. – IR: Bruker IFS 25. – MS: Varian MAT 111 or Varian MAT
311 A. – A Carlo–Erba Fractovap 2900 gas chromatograph with
flame ionization detector and a 10 m ϫ 0.3 mm column coated
with silicon phase OV 101 was used for analytical gas chromato-
graphy. – Preparative thin-layer chromatography was performed on
plates (20 ϫ 20 cm) coated with silica gel or aluminum oxide (60
PF254, Merck).
ESR Experiment: A degassed solution of 7 (0.2 mg, 0.4 µmol) in
n-hexane (1 mL) (4 10–4 mol L–1) was sealed in an ESR sample
tube and irradiated for 5 min at a wavelength of 254 nm. ESR spec-
tra were recorded on a Varian E9 or a Bruker ESP 300 spectro-
meter. The latter instrument was also used for ENDOR measure-
ments.
The ESR spectral simulations were performed using the freeware
program WINSIM.[28]
Scheme 1. Synthesis of trisilane 7 starting from monosilane 14
tert-Butylchlorophenylsilane (15): A solution of tert-butylphenylsil-
ane (14) (26 g, 0.158 mol) in CCl4 (20 mL) was added dropwise to
a suspension of PCl5 (34 g, 0.163 mol) in CCl4 (150 mL) at 0 °C.
After stirring for 2 h at room temperature, the solvent was evapor-
ated in vacuo. In order to remove the salts, the residue was taken-
up in petroleum ether and this solution was filtered under argon
atmosphere. Distillation of the filtrate using a Vigreux column (93–
95 °C/15 Torr) yielded 23.9 g (76%) of a colorless liquid. – IR
(film): ν˜ ϭ 2960 cm–1 (CH), 2160 (SiH), 1460 (tBu), 1360 (tBu),
1120, 830, 820, 740, 700. – 1H NMR (CDCl3): δ ϭ 1.02 (s, 9 H,
tBu), 4.96 (s, 1 H, SiH), 7.3–7.8 (m, 5 H, Ph). – 13C NMR (CDCl3):
δ ϭ 19.4 [SiC(CH3)3], 25.3 [C(CH3)3], 128.0/130.8/131.4/134.4
(Ph). – MS (70 eV); m/z (%): 198 (98) [Mϩ], 183 (4), 155 (43), 141
(100), 120 (12), 105 (51). – HRMS: calcd. for C10H15SiCl (Mϩ)
198.0632; found 198.0684.
Conclusions
The structure of radical 9 follows from the reaction with
ethanethiol, which generates the silacyclobutene 12. This is
further supported by computational analysis of the ESR
spectrum of the radical. In contrast to the parent cyclobut-
enyl/bicyclobut-2-yl radical, where the monocyclic structure
is preferred to the 1,3-bridged bicyclic form,[25] the ‘‘corset
effect’’ caused by the bulky groups should favor the spher-
ical bicyclic isomer of 9. Nevertheless, the calculated geo-
metry of 9 is almost planar, in contrast to the related tert-
butylated cyclobutenylium cations, in which the folding of
the ring is much more marked.[26] The allyl-type radical 9
profits from 1,3-bridging to a lesser extent than would a
comparable cation, but is stabilized by the Si centre.[27]
Moreover, the voluminous tert-butyl and trimethylsilyl sub-
stituents impart a high kinetic stability to 9.
1-tert-Butyl-2,2,2-trimethyl-1-phenyldisilane (16): At room temper-
ature, a solution of 15 (30 g, 0.15 mol) in THF (40 mL) was added
to lithium (5.0 g, 0.72 mol) in THF (350 mL) over a period of 1 h.
In order to ensure complete formation of the silyl anion, the mix-
ture was stirred for a further 10 h. The resulting brown solution
was then added to chlorotrimethylsilane (19 g, 0.175 mol) in THF
1110
Eur. J. Org. Chem. 2000, 1107Ϫ1112

1,2,3,4-Tetra-tert-butyl-4-trimethylsilyl-4-sila-2-cyclobuten-1-yl
(80 mL) over a period of 3 h at 0 °C. After stirring for a further in THF (90 mL) was cooled to –78 °C, whereupon lithium (380 mg,
FULL PAPER
2 h, the mixture was subjected to aqueous work-up. Removal of
55 mmol) was added. After stirring for a further 20 h, the brown
the solvent and vacuum distillation (Vigreux column; 73–75 °C/
solution of the anion was added to a suspension of tri-tert-butyl-
1 Torr) yielded 27.5 g (80%) of 16. – IR (film): ν˜ ϭ 2960 cm–1 (CH), cyclopropenylium tetrafluoroborate (6.5 g, 22 mmol) in THF
2080 (SiH), 1250, 835, 765, 715, 700. – 1H NMR (CDCl3): δ ϭ
0.13 (s, 9 H, SiMe3), 0.98 (s, 9 H, tBu), 3.91 (s, 1 H, SiH), 7.23– resulting mixture was then stirred for a further 24 h and sub-
(100 mL) at –78 °C in 5 mL aliquots over a period of 6 h. The
7.56 (m, 5 H, Ph). – 13C NMR (CDCl3): δ ϭ 0.0 (SiMe3), 18.8
[SiC(CH3)3], 29.0 [C(CH3)3], 127.7/128.6/135.8/136.1 (Ph). – MS
(70 eV); m/z (%): 236 (29) [Mϩ], 179 (90), 135 (88), 121 (63), 105
(76), 83 (100), 73 (96), 59 (33). – HRMS: calcd. for C13H24Si2 (Mϩ)
236.1416; found 236.1407.
sequently allowed to slowly warm to room temperature. After aque-
ous work-up and removal of the solvents, the crude product was
purified by recrystallization from acetone followed by TLC (SiO2,
petroleum ether, 30:50). A second recrystallization from acetone
gave 4.9 g (47%) of pure cyclopropenyltrisilane 7 as a colorless
solid, m.p. 135 °C. – IR (neat): ν˜ ϭ 3075 cm–1 (CH, Ph), 2970
(CH), 1805 (CϭC), 1475 (tBu), 1430 (SiPh), 1390, 1360 (SitBu),
1-Bromo-1-tert-butyl-2,2,2-trimethyldisilane (17): HBr (ca. 40 mL)
was condensed onto phenyldisilane 16 (27 g, 114 mmol) at –196 °C.
After allowing the mixture to warm to –78 °C, stirring was con-
tinued for two days and then the excess HBr was evaporated at
room temperature. Subsequent distillation furnished 25.5 g (94%)
of bromodisilane 17 as a colorless liquid, b.p. 82 °C/22 Torr. – IR
(film): ν˜ ϭ 2970 cm–1 (CH), 2120 (SiH), 1480, 1470 (tBu), 1370,
1255 (SiMe3), 845 (SiMe3), 780, 720. – 1H NMR (CDCl3): δ ϭ 0.27
(s, 9 H, SiMe3), 1.10 (s, 9 H, tBu), 4.15 (s, 1 H, SiH). – 13C NMR
(CDCl3): δ ϭ –1.9 (SiMe3), 24.4 [SiC(CH3)3], 25.4 [C(CH3)3]. – MS
(70 eV); m/z (%): 240 (13) [Mϩ], 238 (12), 183 (6), 181 (6), 73
(100). – HRMS: calcd. for C7H19BrSi2 (Mϩ) 240.0225; found
240.0245.
1
1245 (SiMe3), 1215, 1100, 830 (SiMe3), 810, 735, 700. – H NMR
(CDCl3): δ ϭ 0.30 (s, 9 H, SiMe3), 0.62 (s, 6 H, SiMe2Ph), 0.85/
1.09/1.21/1.32 (4 s, 4 ϫ 9 H, 4 ϫ tBu), 7.22–7.61 (m, 5 H, Ph). –
13C NMR (CDCl3): δ ϭ 4.3/4.4 (diastereotopic SiMe2Ph), 5.7
(SiMe3), 23.1 [SiC(CH3)3], 31.5/31.7/32.6/33.1 [C(CH3)3], 32.1/38.5/
38.9 [C(CH3)3], 129.6/130.2 (CϭC), 127.2/127.8/135.1/143.8 (Ph). –
MS (70 eV); m/z (%): 443 (3) [Mϩ – tBu], 207 (100), 135 (16), 73
(27), 57 (46). – HRMS: calcd. for C26H47Si3 (Mϩ – tBu) 443.2987;
found 443.2997.
2-tert-Butyl-2-(1,2,3-tri-tert-butylcyclopropenyl)-1,1,1-trimethyl-
disilane (11): A degassed solution of silane 7 (310 mg, 0.62 mmol)
in methylcyclohexane (14 mL) was irradiated for 4 h at a wave-
length of 254 nm. The yellow color of 9 that had developed van-
ished upon exposure to oxygen. After removal of the solvent, the
crude product was purified by TLC (SiO2, petroleum ether, 30:50).
From the first band, 140 mg (62%) of pure cyclopropenyldisilane
11 was isolated as a colorless solid, m.p. 68 °C. – IR (film): ν˜ ϭ
2970–2860 cm–1 (CH), 2090 (SiH), 1835 (CϭC), 1480 (tBu), 1465,
2-tert-Butyl-1,1,3,3,3-pentamethyl-1-phenyltrisilane (18): The di-
methylphenylsilyl anion was generated by stirring chlorodimethyl-
phenylsilane (14.5 g, 85 mmol) with lithium pellets (2.0 g,
288 mmol) in THF (200 mL) for 24 h at room temperature. After
removal of the remaining lithium, this solution was added dropwise
to bromodisilane 17 (20.5 g, 85.7 mmol) in THF (100 mL) over a
period of 24 h at –78 °C. The resulting mixture was stirred for a
further 24 h at room temperature and then subjected to aqueous
work-up. Distillation in vacuo (68–70 °C/0.01 Torr) furnished
18.3 g (73%) of trisilane 18. – IR (film): ν˜ ϭ 3080 cm–1 (CH, Ph),
2960 (CH), 2060 (SiH), 1475, 1470 (tBu), 1435 (SiPh), 1370, 1250
(SiMe3), 1110, 860 (SiMe3), 840, 815, 765, 735, 705. – 1H NMR
(CDCl3): δ ϭ 0.05 (s, 9 H, SiMe3), 0.45 (s, 6 H, SiMe2Ph), 0.99 (s,
9 H, tBu), 3.22 (s, 1 H, SiH), 7.28–7.52 (m, 5 H, Ph). – 13C NMR
(CDCl3): δ ϭ –0.3/–0.7 (diastereotopic SiMe2Ph), 1.1 (SiMe3), 19.5
[SiC(CH3)3], 31.5 [C(CH3)3], 127.7/128.4/134.0/140.9 (Ph). – MS
(70 eV); m/z (%): 294 (4) [Mϩ], 279 (30), 237 (33), 220 (63), 164
(47), 135 (60), 73 (100). – HRMS: calcd. for C11H21Si3 (Mϩ – tBu)
237.0951; found 237.0901.
1
1395, 1365, 1250, 860, 835, 770, 730. – H NMR (C6D6): δ ϭ 0.50
(s, 9 H, SiMe3), 1.12/1.30/1.36/1.40 (4 s, 4 ϫ 9 H, 4 ϫ tBu), 3.48 (s,
1 H, SiH). – 13C NMR (C6D6): δ ϭ 7.8 (SiMe3), 21.6 [SiC(CH3)3],
31.2/31.4/32.1/32.2 [C(CH3)3], 31.7/31.9/36.5/37.9 [C(CH3)3
ϩ
bridgehead], 128.9/129.8 (CϭC). – MS (70 eV); m/z (%): 351 (5)
[Mϩ], 309 (6), 207 (100), 73 (43), 57 (41), 41 (22). – HRMS: calcd.
for C21H43Si2 (Mϩ) 351.2903; found 351.2879.
2,3,4-Tri-tert-butyl-4-trimethylsiloxyl-4-oxa-5-silacyclopent-2-en-1-
one (13): The TLC separation described above for the isolation of
11 was followed by a second TLC (Al2O3, petroleum ether, 30:50)
of the residue remaining after evaporation of the solvent from the
other fractions. The yield of 13 was 9%. – IR (neat): ν˜ ϭ 2958
cm–1 (CH), 1744 (CϭO), 1581 (CϭC), 1474 (tBu), 1393, 1364
1
2-tert-Butyl-2-chloro-1,1,3,3,3-pentamethyl-1-phenyltrisilane (19):
In order to prepare chlorotrisilane 19, trisilane 18 (1.38 g,
4.7 mmol) was dissolved in tetrachloromethane (10 mL) and the
resulting solution was stirred at 50 °C. After 20 h, the reaction was
complete and all volatile materials were removed in vacuo. The
remaining colorless liquid (b.p. 100–105 °C/0.04 Torr) was used
without further purification. – IR (film): ν˜ ϭ 3070 cm–1 (CH, Ph),
2950 (CH), 1470, 1460 (tBu), 1430 (SiPh), 1400, 1360, 1250
(SiMe3), 1120, 1110, 1070, 840, 815, 790, 735, 700. – 1H NMR
(CDCl3): δ ϭ 0.08 (s, 9 H, SiMe3), 0.49/0.50 (2 s, 2 ϫ 3 H, diastere-
otopic SiMe2Ph), 0.96 (s, 9 H, tBu), 7.32–7.56 (m, 5 H, Ph). –
13C NMR (CDCl3): δ ϭ –2.2/–2.1 (diastereotopic SiMe2Ph), –0.5
(SiMe3), 23.1 [SiC(CH3)3] 28.0 [C(CH3)3], 127.9/129.0/134.2/138.3
(Ph). – MS (70 eV); m/z (%): 328 (8) [Mϩ], 313 (4), 271 (9), 220
(59), 164 (49), 135 (100), 73 (100). – HRMS: calcd. for C11H20ClSi3
(Mϩ – tBu) 271.0560; found 271.0573.
(SitBu), 1254 (SiMe3), 1187, 1063 (SiO), 848 (SiMe3). – H NMR
(CDCl3): δ ϭ 0.13 (s, 9 H, SiMe3), 1.00 (s, 9 H, SitBu), 1.36 (s, 9
H, 2-tBu), 1.41 (s, 9 H, 3-tBu). – 13C NMR (CDCl3): δ ϭ 1.45
(SiMe3), 18.6 [SiC(CH3)3], 26.5 [SiC(CH3)3], 31.8/33.4 [C(CH3)3],
34.3/37.0 [C(CH3)3], 155.2/161.3 (CϭC), 168.0 (CϭO). – 29Si NMR
(CDCl3): δ ϭ –17.3 (SiMe3), 13.5 (ring). – MS (70 eV); m/z (%):
356 (22) [Mϩ], 341 (21), 313 (15), 299 (19), 215 (30), 133 (90), 73
(28), 57 (100). – HRMS: calcd. for C18H36Si2O3 (Mϩ) 356.2203;
found 356.2192.
1,2,3,4-Tetra-tert-butyl-3-trimethylsilyl-3-silacyclobut-1-ene (12): A
solution of 7 (268 mg, 0.53 mmol) in methylcyclohexane (20 mL)
was irradiated for 2.5 h at room temperature with light of wave-
length 254 nm. Addition of ethanethiol (0.5 mL) caused the solu-
tion to turn colorless. After removal of the solvents, purification of
the residue by TLC (SiO2, pentane) yielded a mixture of the H-
silanes 11 and 12. Subsequent preparative HPLC (Lichrosorb RP-
18, Merck) furnished pure silacyclobutene 12 (5%) as a colorless
solid, m.p. 70 °C. – IR (neat): ν˜ ϭ 3023 cm–1 (CH), 1517 (Cϭ
2-tert-Butyl-2-(tri-tert-butylcyclopropenyl)-1,1,3,3,3-pentamethyl-1-
phenyltrisilane (7): A solution of chlorotrisilane 19 (6.9 g, 21 mmol)
Eur. J. Org. Chem. 2000, 1107Ϫ1112
1111

G. Maier, A. Kratt, A. Schick, H. P. Reisenauer, F. Barbosa, G. Gescheidt
FULL PAPER
[10]
Aryl tri-tert-butylsilacyclobutadienes can be prepared by this
C), 1466 (tBu), 1387, 1360 (SitBu), 1245 (SiMe3), 1103, 1009, 850
(SiMe3). – H NMR (CDCl3): δ ϭ 0.27 (s, 9 H, SiMe3), 1.03/1.06/
[10a]
route and have been extensively studied by Fink et al.:
M.
1
J. Fink, D. B. Puranik, M. P. Johnson, J. Am. Chem. Soc. 1988,
1.20/1.24 (4 s, 4 ϫ 9 H, 4 ϫ tBu), 2.08 (s, 1 H, CH). – 13C NMR
(CDCl3): δ ϭ 3.4 (SiMe3), 21.7 [SiC(CH3)3], 30.1/31.4/32.8/33.1
[C(CH3)3], 31.9/33.8/37.4 [C(CH3)3], 49.9 [C(4)], 161.2/164.4 (Cϭ
C). – MS (70 eV); m/z (%): 366 (2.4) [Mϩ], 351 (3), 309 (73), 293
(37), 240 (20), 171 (12), 73 (100), 57 (61). – HRMS: calcd. for
C22H46Si2 (Mϩ) 366.3138; found 366.3186.
110, 1315–1316. – [10b] D. B. Puranik, M. J. Fink, J. Am. Chem.
[10c]
Soc. 1989, 111, 5951–5952. –
J. R. Gee, W. A. Howard,
G. L. McPherson, M. J. Fink, J. Am. Chem. Soc. 1991, 113,
5461–5462.
[11]
[12]
H. Vancik, G. Raabe, M. J. Michalczyk, R. West, J. Michl, J.
Am. Chem. Soc. 1985, 107, 4097–4098.
M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J.
Am. Chem. Soc. 1985, 107, 3902. AM1 calculations were per-
formed using MOPAC 6.0: J. J. P. Stewart, F. J. Seiler Res. Lab.
U.S. Air Force Academy, Colorado Springs, 1994.
M. S. Gordon, T. J. Barton, H. Nakano, J. Am. Chem. Soc.
1997, 119, 11966.
M. Kraftory, M. Kapon, M. Botoshansky: ‘‘The Structural
Chemistry of Organosilicon Compounds’’ in: S. Patai ‘‘The
Chemistry of Organosilicon Compounds’’, Vol. 2 (Eds.: Y. Ape-
loig, Z. Rappaport), John Wiley, Chichester, 1998, 181–265.
W. Kohn, A. D. Becke, A. G. Parr, J. Phys. Chem. 1996, 100,
12974.
R. G. Parr, W. Yang, Density Functional Theory of Atoms and
Molecules, Oxford University Press, Oxford, 1989.
J. W. Gauld, L. A. Eriksson, L. Radom, J. Phys. Chem. A 1997,
101, 1352.
R. Batra, B. Giese, M. Spichty, G. Gescheidt, K. N. Houk, J.
Phys. Chem 1996, 100, 18371.
P. J. Stephens, J. F. Devlin, F. Chabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623; A. D. Becke, Phys. Rev. A 1988,
38, 3098; A. D. Becke, J. Chem. Phys. 1993, 98, 1372.
H. Bock, B. Roth, G. Maier, Angew. Chem. 1980, 92, 213–214;
Angew. Chem. Int. Ed. Engl. 1980, 19, 209–210.
F. Gerson, Hochauflösende ESR-Spektroskopie, Verlag Chemie,
Weinheim, 1967.
H. Kurreck, B. Kirste, W. Lubitz, Electron Nuclear Double Res-
onance Spectroscopy of Radicals in Solution, VCH, Weinheim,
1988.
A. Boudin, G. Cerveau, C. Chuit, R. J. P. Corriu, Bull. Chem.
Soc. Jpn. 1988, 61, 101–106.
U. Wessolek-Kraus, Ph. D. Dissertation, University of
Giessen, 1988.
Acknowledgments
[13]
[14]
We wish to thank the Swiss National Science Foundation, the
Deutsche Forschungsgemeinschaft, and the Fonds der Chemischen
Industrie for their support.
[1]
[15]
[16]
[17]
[18]
[19]
G. Maier, Angew. Chem. 1988, 100, 317–341; Angew. Chem. Int.
Ed. Engl.1988, 27, 309–332.
[2]
G. Maier, F. Fleischer, Liebigs Ann. 1995, 169–172; G. Maier,
F. Fleischer, Tetrahedron Lett. 1991, 31, 57–60.
[3]
G. Maier, F. Fleischer, H.-O. Kalinowski, Liebigs Ann. 1995,
173–186.
[4] [4a]
G. Maier, D. Born, I. Bauer, R. Wolf, R. Boese, D. Cremer,
[4b]
Chem. Ber. 1994, 127, 173–189. –
G. Maier, R. Wolf, H.-O.
Kalinowski, R. Boese, Chem. Ber. 1994, 127, 191–200.
G. Maier, R. Wolf, H.-O. Kalinowski, Chem. Ber. 1994, 127,
201–204.
[5]
[6] [6a]
[20]
[21]
[22]
P. Lingelbach, Ph. D. Dissertation, University of Giessen,
[6b]
1989. –
A. Schick, Ph. D. Dissertation, University of
[6c]
Giessen, 1992. –
A. Kratt, Ph. D. Dissertation, University
of Giessen, 1998. – Summary: [6d] G. Maier, A. Meudt, J. Jung,
H. Pacl, ‘‘Matrix Isolation Studies of Silicon Compounds’’ in:
S. Patai ‘‘The Chemistry of Organosilicon Compounds’’, Vol. 2
(Eds.: Y. Apeloig, Z. Rappaport), John Wiley, Chichester, 1998,
1143–1185.
[23]
[24]
[25]
[26]
[27]
[28]
[7]
M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A.
Petersson, J. A. Montgomery, K. Raghavachari, M. A. Al-La-
ham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslow-
ski, B. B. Stefanov, A. Nanayakkara, M. Challacombe, C. Y.
Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S.
Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley,
D. J. Defrees, J. Baker, J. J. P. Stewart, M. Head-Gordon, C.
Gonzalez, J. A. Pople, Gaussian 94, Revision E-2, Gaussian,
Inc., Pittsburgh PA, 1995.
G. Maier, H. P. Reisenauer, J. Jung, H. Pacl, H. Egenolf, Eur.
J. Org. Chem. 1998, 1297–1305 and refs. cited therein.
G. W. Schriver, M. J. Fink, M. S. Gordon, Organometallics
1987, 6, 1977–1984.
P. J. Krusic, J. P. Jesson, J. K. Kochi, J. Am. Chem. Soc. 1969,
91, 4566–4568.
G. Maier, R. Emrich, K.-D. Malsch, K.-A. Schneider, M.
Nixdorf, H. Irngartinger, Chem. Ber. 1985, 118, 2798–2810.
T. Kawamura, P. Meakin, J. K. Kochi, J. Am. Chem. Soc. 1972,
94, 8065–8072.
[8]
[9]
D. R. Duling, PEST WINSIM (NIEHS, Research Triangle
Park, NC, USA, 1995).
Received July 26, 1999
[O99455]
1112
Eur. J. Org. Chem. 2000, 1107Ϫ1112