Synthesis of 1,4-disilacyclohexa-2,5-dienes

Article abstract of DOI:10.1016/S0022-328X(02)01142-7

Title compounds of the type 2,3,5,6-tetraphenyl-1,4-di-X-1,4-di-Y-1,4-disilacyclohexa-2,5-diene wherein X = Y = NMe₂ (4); X = NMe₂, Y = Cl (cis, trans-5); X = NMe₂, Y = Me [(trans)-6] and X = t-Bu, Y = Cl (trans-8) were synthesized from Si₂(NMe₂)₅Cl, sym-Si₂(NMe₂)₄Cl₂, sym-Si₂(NMe₂)₄Me₂, and sym-Si₂Cl₄(t-Bu)₂, respectively, in the presence of diphenylacetylene at 200 °C. Similarly the analogous title compound from the combination of 1-phenyl-1-propyne and Si₂(NMe₂)₅Cl [X = Y = NMe₂ (cis and trans-7) was synthesized. In all cases where cis/trans diastereomers could arise from two different silicon substituents (5, 6, 8) the trans isomer was the sole or dominant product. Evidence for the intermediacy of the silylene Si(NMe₂)₂ in these reactions was gained from a trapping experiment. Compound 4 upon treatment with SiCl₄, SiBr₄ or PI₃ provided the corresponding 1,1,4,4-tetrahalo derivatives 9a-c, respectively. Treatment of 4 with MeOH or PhOH gave the 1,1,4,4-tetramethoxy and tetraphenoxy analogues 9d and 9e, respectively. The tetrachloro derivative 9a upon LAH reduction led to the corresponding tetrahydro compound 10, while the reaction of 9a with H₂O gave the tetrahydroxy derivative 11. Allowing (trans)-6 to react with SiCl₄ provided a ca. 1:1 cis/trans ratio of the derivative 12 in which X = Cl, Y = Me, and possible pathways that rationalize this loss of stereochemistry are proposed. Synthesis of trans-13 in which X = t-Bu, Y = H was achieved by LAH reduction of 8. All of the title compounds except 8 experience free phenyl rotation at room temperature. At - 30 °C this rotation in 8 is essentially halted. The molecular structures of 4, 8, 9c, 9e, 10 and 13 were determined by X-ray crystallography.

Full text of DOI:10.1016/S0022-328X(02)01142-7

Journal of Organometallic Chemistry 649 (2002) 276–288
www.elsevier.com/locate/jorganchem
Synthesis of 1,4-disilacyclohexa-2,5-dienes
Jinchao Yang, Ilia Guzei, John G. Verkade *
Chemistry Department, College of Liberal Arts and Sciences, Iowa State Uni6ersity, Gilman Hall, Ames, IW 50011-3111, USA
Received 27 August 2001; received in revised form 7 December 2001; accepted 7 December 2001
Dedication to Academician M.G. Voronkov on the occasion of this 80th birthday
Abstract
Title compounds of the type 2,3,5,6-tetraphenyl-1,4-di-X-1,4-di-Y-1,4-disilacyclohexa-2,5-diene wherein X=Y=NMe2 (4);
X=NMe2, Y=Cl (cis, trans-5); X=NMe2, Y=Me [(trans)-6] and X=t-Bu, Y=Cl (trans-8) were synthesized from
Si2(NMe2)5Cl, sym-Si2(NMe2)4Cl2, sym-Si2(NMe2)4Me2, and sym-Si2Cl4(t-Bu)2, respectively, in the presence of diphenylacetylene
at 200 °C. Similarly the analogous title compound from the combination of 1-phenyl-1-propyne and Si2(NMe2)5Cl [X=Y=
NMe2 (cis and trans-7) was synthesized. In all cases where cis/trans diastereomers could arise from two different silicon
substituents (5, 6, 8) the trans isomer was the sole or dominant product. Evidence for the intermediacy of the silylene Si(NMe2)2
in these reactions was gained from a trapping experiment. Compound 4 upon treatment with SiCl4, SiBr4 or PI3 provided the
corresponding 1,1,4,4-tetrahalo derivatives 9a–c, respectively. Treatment of 4 with MeOH or PhOH gave the 1,1,4,4-tetramethoxy
and tetraphenoxy analogues 9d and 9e, respectively. The tetrachloro derivative 9a upon LAH reduction led to the corresponding
tetrahydro compound 10, while the reaction of 9a with H2O gave the tetrahydroxy derivative 11. Allowing (trans)-6 to react with
SiCl4 provided a ca. 1:1 cis/trans ratio of the derivative 12 in which X=Cl, Y=Me, and possible pathways that rationalize this
loss of stereochemistry are proposed. Synthesis of trans-13 in which X=t-Bu, Y=H was achieved by LAH reduction of 8. All
of the title compounds except 8 experience free phenyl rotation at room temperature. At −30 °C this rotation in 8 is essentially
halted. The molecular structures of 4, 8, 9c, 9e, 10 and 13 were determined by X-ray crystallography. © 2002 Published by
Elsevier Science B.V.
Keywords: 1,4-Disilacyclohexadiene synthesis; Disilane reactions; Acetylene reactions
Thus, it appeared to us that disilanes with multiple
dimethylamino groups could be quite efficient for syn-
1. Introduction
thesizing 1,4-disilacyclohexa-2,5-dienes under compara-
We reported in a recent communication [1] that
tively mild conditions, in sharp contrast to the original
disilanes Si2Cl6 and Si2(OMe)6 are good catalysts for
synthesis of 1 which was reported by others to take
cyclo-trimerizing alkynes into the corresponding aro-
matic products in moderate to excellent yields at ca.
place at ca. 450 °C where the silylene diradical SiCl2
(generated from Si2Cl6 via disproportionation) was pro-
200 °C in a sealed tube. A reaction pathway involving
posed as the reactive intermediate [3]. Although gas
silyl radicals was proposed [1], which was further sup-
phase reactions of acetylenes with
ported with additional experiments in a more recent
publication [2] from these laboratories. We also dis-
closed that in reactions with alkynes increased substitu-
tion of the chloro groups of Si2Cl6 with dimethylamino
silylene sources at elevated temperatures were for many
substituents gave generally increasing ratios of 1,4-disi-
years a traditional route to substituted 1,4-disilacyclo-
lacyclohexa-2,5-dienes to cyclotrimerized products.
hexa-2,5-dienes [4], other approaches have since been
developed in which transition complexes of metals such
as Ni or Pd were utilized as catalysts, allowing reac-
tions to be carried out at lower temperatures more
quickly [5–7]. An efficient catalytic process that takes
* Corresponding author. Tel.: +1-515-294-5023; fax: +1-515-294-
0105.
E-mail address: jverkade@iastate.edu (J.G. Verkade).
0022-328X/02/$ - see front matter © 2002 Published by Elsevier Science B.V.
PII: S0022-328X(02)01142-7

J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
277
place in solution at 120 °C to give 2a–f was more
recently described [5], but it uses trisilanes as the source
of silylene (reaction 1) which are less accessible than
disilanes. Other less common methods include Pt-cata-
lyzed hydrosilylation–dimerization of acetylenes that
contain chelating substituents, and the silylation–cy-
clization of bis(organolithium) derivatives (reaction 2)
[6,7]. A few years ago, a new synthetic approach was
reported to give bis(1,4-disilacyclohexa-2,5-dienes 3a–c
via a bis(cyclodisilabutane)/alkyne reaction catalyzed
by (H2CꢀCH2)Pt(PPh3)2 (Scheme 1) [8]. These ap-
proaches, however, did not provide us with a practical
way to synthesize 1,4-disilacyclohexa-2,5-dienes bearing
functional groups that are easily transformed for fur-
ther elaboration.
Scheme 1.
In this paper we describe syntheses for 4–8, 9a–e
and 10–13 in good yields under relatively mild reaction
conditions. Reaction pathways for the formation of
these products are also suggested. The molecular struc-
tures of 4, 8, 9c, 9e, 10 and 13 were obtained by X-ray
crystallography.
(1)
(2)
2. Experimental
NMR spectra were measured on Bruker AX-200
(29Si), Bruker DRX-400 (1H and 13C), and Bruker
MSL-300 (solid MAS 29Si) instruments. High resolution
mass spectra (HRMS) and X-ray structure determina-
tions were performed in the Iowa State University’s
Instrument Services unit and elemental analysis were
carried out by Desert Analytics.
Diethyl ether, pentane and THF were distilled from
sodium–benzophenone; CHCl3 was distilled from CaH2
under nitrogen; and xylenes were distilled under N2
X
Y
6
R
6
R%
R%%
Ph
Ph
Ph
Me
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
4
NMe2 NMe2 Ph
NMe2 Cl
NMe2 Me
NMe2 NMe2 Me
NMe2 NMe2 Ph
t-Bu
Cl
Ph
Ph
Ph
Ph
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
cis,trans-5
(trans)-6
trans-7
cis-7
trans-8
9a
9b
9c
9d
9e
Ph
Ph
,
from sodium and stored over 4 A molecular sieves.
Hexachlorodisilane, symmetrical-dimethyltetrachloro-
disilane and hexamethyldislane were purchased from
Aldrich Chemical Co., and hexamethoxydisilane was
purchased from Gelest, Inc. and were used as received.
Benzil, 1-phenyl-1,2-propanedione, anhydrous dimethy-
lamine, tert-butylchlorodiphenylsilane, silicon tetra-
Cl
Cl
Br
I
OMe
OPh
H
OH
Cl
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Br
I
OMe
OPh
H
OH
Me
H
chloride,
phosphorus
trichloride,
phosphorus
10
11
tribromide, silicon tetrabromide, phosphorus tri-iodide,
diphenylacetylene, phenylacetylene, 1-phenyl-1-propyne
and sodium–mercury amalgam were purchased from
Aldrich Chemical Co. and were also used as received.
cis,trans-12
trans-13
t-Bu

278
J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
2.1. Synthesis of disilane starting materials
After cleaning the inside wall of the tube with a cotton
swab, the tube was flushed with nitrogen and then it
was cooled with liquid nitrogen (for liquid reaction
mixtures) and flame sealed under vacuum. In these
reactions, the height of the reaction mixture was about
one fifth of that of the glass tube. The mixture in the
glass tube was heated to ca. 200 °C for 2 days in an oil
bath. Crystal formation was generally observed after
the reaction mixture was allowed to cool to ambient
temperature. The tube was quickly opened in a hood,
and the liquid portion decanted. The crystalline
product was washed several times by the stated amount
of pentane, and was dried under vacuum.
Dimethyl-tetrakis(dimethylamino)disilane [10], sym-
metrical-di-tert-butyltetraphenyldisilane [11], and sym-
metrical-di-butyltetrachlorodisilane [11] were prepared
by literature methods. Although the synthesis of hexak-
is(dimethylamino)disilane,
pentakis(dimethylamino)-
chlorodisilane and tetrakis(dimethylamino)dichloro-
disilane had been reported [9,12], we used a modified
method described below that gave higher yields and
purities. The following example describes the general
procedure for the synthesis of disilanes bearing multiple
dimethylamino groups using the improved method.
2.1.1. Synthesis of symmetrical-di-tert-butyldichloro-
bis(dimethylamino)disilane
2.1.3. Synthesis of 2,3,5,6-tetraphenyl-1,1,4,4-
tetrakis(dimethylamino)-1,4-disila cyclohexa-2,5-diene
In a 250 ml flask equipped with a magnetic stirrer,
Et2O (ca. 100 ml) was added, followed by the addition
of symmetrical-di-tert-butyltetrachlorodisilane (3.12 g,
10.0 mmol). The flask was stoppered with a septum,
and its weight was recorded before dimethylamine was
introduced. The flask was then cooled with liquid nitro-
gen, and dimethylamine was introduced from a gas
cylinder through a needle, whereupon a white precipi-
tate formed immediately. The flask was weighed fre-
quently and the addition of dimethylamine was stopped
when 2.70 g (60.0 mmol, 50% excess) had been intro-
duced. (For the syntheses of sym-Si2Cl2(NMe2)4 and
Si2Cl(NMe2)5, however, the weight of HNMe2 was
carefully controlled so that only the stoichiometric
amount of dimethylamine was added.) The mixture was
allowed to warm up slowly and was vigorously stirred
for 20 h at room temperature. After the reaction,
unreacted HNMe2 was allowed to escape through a
needle inserted into the septum. The solid material
(presumably [H2NMe2]Cl) was filtered off and washed
with 5×30 ml of Et2O. The solvent and the volatile
materials were removed under reduced pressure giving
the product (85% yield) which in the case of sym-
Sit2Bu2Cl2(NMe2)2 was analyzed as a mixture of two
conformers (in a 1:1 ratio) that possessed satisfactory
purity as determined by 1H- and 13C-NMR spec-
(4)
Pentakis(dimehthylamino)chlorodislane (3.11 g, 10.0
mmol) was combined under nitrogen with dipheny-
lacetylene (2.05 g, 11.5 mmol) in a tube which was
sealed under vacuum and heated to 200 °C for 2 days
in an oil bath, during which time the color of the
reaction mixture became brown. Upon allowing the
tube to cool undisturbed to room temperature, crystal
formation was observed. The tube was opened in a
hood and the liquid portion was removed by decanta-
tion. The crystalline material was washed with a 5×10
ml of pentane. Drying the crystals under vacuum af-
1
forded 2.64 g of the title product (45% yield). H-NMR
(CDCl3): l 6.83–7.03 (m, 20H), 2.34 (s, 24H); 13C-
NMR (CDCl3): l 158.43, 143.69, 128.88, 126.95,
124.76, 38.11; 29Si-NMR (MAS): l −33.11. M.p.:
260–265 °C. CIMS (NH3, negative); m/z (relative in-
tensity, ion): 588.2 ([M−], 18.83), 544.2 ([M−NMe2],
4.04). Elemental analysis: Calc. for C36H44N4Si2: C,
73.42; H, 7.53; N 9.51. Found: C, 72.40; H, 7.53; N
9.39%. The liquid product contained a substantial
amount of Si(NMe2)3Cl according to H, 13C and mass
spectroscopies. Crystals of 4 suitable for X-ray analysis
were obtained directly from the product of the reaction.
1
1
troscopy. H-NMR (CDCl3): l 2.65 (s, 12H), 2.61 (s,
2.1.4. Synthesis of isomeric-2,3,5,6-tetraphenyl-1,4-
bis(dimethylamino)-1,4-dichloro-1,4-disilacyclo-
hexa-2,5-dienes (5)
12H), 1.04 (s, 18H), 1.01 (s, 18H); 13C-NMR (CDCl3):
l 39.32, 38.70, 26.90, 26.56, 26.38, 26.22. M.p.: 25 °C.
EIHRMS; Calc. for C12H30N2Cl2Si2: 328.1325; found:
328.1328. Elemental analysis: Calc. for C12H30N2Cl2Si2:
C, 43.90; H, 9.14; Cl, 21.65; N, 8.54; Si, 17.07; Found:
C, 43.47; H, 9.09; Cl, 21.38; N, 8.12; Si, 17.98%. The
yields of Si2(NMe2)6 and Si2Cl(NMe2)5 were 80 and
85%, respectively.
The conditions for this reaction were similar to those
for the synthesis of 4 except that Si2Cl2(NMe2)4 (3.03 g,
10.0 mmol) and diphenylacetylene (2.07 g, 11.5 mmol)
were used as starting materials. After the reaction, the
product was analyzed by H- and 13C-NMR spectro-
1
scopies as a mixture of two isomeric 1,4-disila-2,5-di-
enes (with one major isomer and a trace amount of the
other) and a minor amount of hexaphenylbenzene. The
by-product [Si(NMe2)3Cl], was separated by decanta-
tion and the crystalline material was washed with 5×
10 ml of pentane. Drying the crystals under vacuum
2.1.2. General method for synthesizing
1,4-disilacyclohexa-2,5-dienes
The disilane and an alkyne (diphenylacetylene or
1-phenyl-1-propyne) were combined in a glass tube.

J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
279
afforded 1.43 g of material which was analyzed by
1H-NMR spectroscopy as 75% of a mixture of isomers
of 5 and 25% hexaphenylbenzene. Removal of the
hexaphenylbenzene was accomplished by recrystalliza-
tion from methylene chloride, giving a 10% yield of 5.
Attempts to separate the isomers by column chro-
by drying under vacuum. The product analyzed as a
mixture of about 95% of one of the isomers with
presumably trans-7 isomer dominating, judging from
1
the single H-NMR peak and the single 13C-NMR peak
for the NMe2 group for the dominant species. Yield:
1
70%. H-NMR (CDCl3): l 6.94–7.29 (m, 10H); 13C-
1
matography failed. H-NMR (CDCl3): l 7.00–7.08 (m,
NMR (CDCl3): l 156.25, 152.28, 144.15, 127.90,
127.69, 125.07, 37.80, 18.59. M.p.: 168–170 °C.
EIHRMS; Calc. for C26 H40N4Si2: 464.2792; found:
464.2798. This compound was quite soluble in pentane,
in contrast to the other 1,4-disilacyclohexa-2,5-dienes
synthesized here. That this product is an isomeric mix-
ture is supported by its reaction with SiCl4 which
produced products consisting of 95% of one tetrachloro
product. 1H-NMR (CDCl3): l 7.16–7.44 (m, 10 H),
1.91 (s, 6H); 13C-NMR (CDCl3): l 153.63, 152.22,
136.72, 128.52, 128.16, 127.59, 17.90.
20H), 2.33 (s, 12H); 13C-NMR (CDCl3): l 157.30,
140.19, 128.76, 127.33, 125.97, 37.23. M.p. (dec.):
250 °C. CIMS (NH3, positive); m/z (relative intensity,
ion): 571.1 ([M+], 20.13), 535.1 ([M−Cl]+, 6.07, 525.9
([M−NMe2]+,
3.55).
EIHRMS;
Calc.
for
C32H32Cl2N2Si2: 570.1481; found: 570.1493.
2.1.5. Synthesis of 2,3,5,6-tetraphenyl-1,4-dimethyl-1,4-
bis(dimethylamino)-1,4-disilacyclohexa-2,5-diene (6)
In a glass tube were placed sym-Si2Me2(NMe2)4 (2.62
g, 10.0 mmol) and then diphenylacetylene (2.07 g, 11.5
mmol). The tube was cooled in liquid nitrogen, flamed
sealed under vacuum and heated to 200 °C for 2 days,
during which time the reaction mixture was observed to
become light brown. After the reaction, the tube was
allowed to cool to room temperature upon which crys-
tals were observed to form. The tube was opened in a
hood, the liquid portion was removed by decantation
and the crystals were washed with 5×10 ml of pentane.
Drying the crystals under vacuum gave (presumably
2.1.7. Synthesis of trans-2,3,5,6-tetraphenyl-1,4-
dichloro-1,4-di-tert-butyl-1,4-disilacyclohexa-2,5-diene
(trans-8)
Symmetrical-di-tert-butyldichlorobis(dimethylamino)-
disilane (3.29 g, 10.0 mmol) was combined with
diphenylacetylene (2.13 g, 12.0 mmol) in a glass tube
which was then cooled with liquid nitrogen and flame
sealed under vacuum. The tube was heated to 210 °C
for 3 days, during which time a brown solution formed.
The tube was allowed to cool to ambient temperature,
and was further cooled at ca. −18 °C in a freezer for
2 days. However, no crystal formation was observed.
The tube was opened in a hood, and pentane [precooled
in dry ice (20 ml)] was added to precipitate the product.
The tube was then stoppered with a septum and stored
in a freezer for 5 h, during which time the product
precipitated. The solution was removed with a syringe,
and the precipitate was washed with 5×5 ml of pen-
tane. Drying the precipitate under vacuum afforded the
product as a powder. Only one isomer (trans-8) was
1
trans) 6 in 75% yield. H-NMR (CDCl3): l 6.82–7.06
(m, 20H), 2.38 (s, 12H), 0.12 (s, 6H); 13C-NMR
(CDCl3): l 157.40, 143.29, 128.34, 127.08, 124.85,
38.41, −3.36. M.p.: 243–245 °C. EIHRMS; Calc. for
C34H38N2Si2: 530.2574; found: 530.2579. Elemental
analysis: Calc. for C34H38N2Si2: C, 76.98; H, 7.17; N,
5.28, Si, 10.57. Found: C, 77.00; H, 7.26; N, 5.28; Si,
10.75%. Although the crystals obtained were suitable
for X-ray diffraction, the molecular structure could not
be solved (see Section 3).
1
observed. Yield: 5%. H-NMR (CDCl3): l 6.52–7.54
2.1.6. Synthesis of 3,6-dimethyl-2,5-diphenyl-1,1,4,4-
tetrakis(dimethylamino)-1,4-disilacyclohexa-2,5-diene
(trans-7) and 3,5-dimethyl-2,6-diphenyl-1,1,4,4-
tetrakis(dimethylamino)-1,4-disilacyclohexa-2,5-diene
(cis-7)
Pentakis(dimethylamino) chlorodisilane (3.11 g, 10.0
mmol) was combined with 1-phenyl-1-propyne (1.11 g,
10.5 mmol) in a glass tube which was then cooled with
liquid nitrogen and flame sealed under vacuum. The
tube was heated to 200 °C for 2 days in an oil bath
during which time the reaction mixture (a brown solu-
tion) was observed to reflux. After the reaction, the
tube was allowed to cool to room temperature followed
by further cooling to −18 °C in a freezer for 1 day.
The tube was then opened in a hood, the liquid portion
was removed by decantation and the solid was washed
with 3×10 ml of pentane precooled in dry ice, followed
(m, br, 20H), 0.51 (s, 18H); 13C-NMR (CDCl3): l
156.95, 141.79, 130.01, 127.18 (br), 125.99, 26.70, 22.27.
M.p.: 285–287 °C. EIHRMS; Calc. for C36H38Cl2Si2:
596.1889; found: 596.1898. Elemental analysis: Calc. for
C36H38Cl2Si2: C, 72.48; H, 6.29; Cl, 11.91; Si, 9.40;
found: C, 72.26; H, 6.33; Cl, 11.94; Si, 9.01%. Crystals
suitable for X-ray diffraction spectroscopy were ob-
tained by recrystallization from CHCl3.
2.1.8. General procedure for the synthesis of 2,3,5,6-
tetraphenyl-1,1,4,4-tetrahalo-1,4-disilacyclohexa-2,5-
dienes 9a–c
Compound 4 was halogenated with SiCl4, PCl3, SiBr4
or PI3 to obtain the corresponding title products in 95,
95, 95 and 70% yields, respectively. The synthesis of 9a
is given as an example. In a 100 ml flask equipped with

280
J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
a magnetic stirrer, were placed 4 (2.94 g, 5.00 mmol)
and then Et2O (50 ml). SiCl4 (8.50 g, 50.0 mmol) was
introduced through a syringe and the mixture was
refluxed at 45 °C for 24 h during which time a white
precipitate formed. After the reaction, the mixture was
filtered under nitrogen and the white precipitate was
washed with 5×10 ml of pentane. Drying the precipi-
tate under vacuum afforded the product as a powder.
784.2475. Elemental analysis: Calc. for C52H40O4Si2: C,
79.59; H, 5.10. Found: C, 79.10; H, 5.19%. Crystals
suitable for X-ray diffraction spectroscopy were ob-
tained by recrystallization from CHCl3.
2.1.13. Synthesis of 2,3,5,6-tetraphenyl-1,4-
disilacyclohexa-2,5-diene (10)
In a 50 ml flask equipped with a magnetic stirrer
were placed 9a (2.77 g, 5.00 mmol) and then THF (25
ml and LiAlH4 (0.76 g, 20 mmol). The mixture was
refluxed at 75 °C for 24 h. After the reaction, LiAlH4
was destroyed with 20 ml of MeOH. The mixture was
filtered, and volatiles were removed under reduced pres-
sure. The product was purified by column chromatog-
1
Yield: 95%. H-NMR (CDCl3): l 7.07–7.18 (m, 20H);
13C-NMR (CDCl3): l 155.96, 136.71, 128.93, 127.17.
M.p. (dec.): 287 °C. EIHRMS; Calc. for C28H20Cl4Si2:
553.9828; found: 553.9832. Elemental analysis: Calc. for
C28H20Cl4Si2: C, 60.64; H, 3.61; Cl, 25.63; Si 10.11;
found: C, 60.60; H, 3.54; Cl, 25.47; Si, 10.04%.
1
raphy using CHCl3 as the eluent. Yield: 70%. H-NMR
2.1.9. Characterization of 2,3,5,6-tetraphenyl-1,1,4,4-
tetrabromo-1,4-disilacyclohexa-2,5-diene (9b)
(CDCl3): l 7.01–7.16 (m, 20H), 4.58 (s, 4H); 13C-NMR
(CDCl3): l 152.01, 141.67, 128.14, 128.06, 126.34. M.p.:
240–242 °C. EIHRMS; Calc. for C28H24Si2: 416.1417;
found: 416.1422. Elemental analysis: Calc. for
C28H24Si2: C, 80.77; H, 5.77; Si, 13.46; found: C, 80.55;
H, 5.49; Si, 13.15%. Subliming the compound under
vacuum at 200 °C gave crystals suitable for X-ray
diffraction.
1H-NMR (CDCl3): l 7.07–7.20 (m, 20H); 13C-NMR
(CDCl3): l 155.13, 136.70, 129.26, 127.51, 127.12. M.p.:
280 °C (decomposition). EIHRMS; Calc. for
C28H20Br4Si2: 729.7827; found: 729.7828.
2.1.10. Characterization of 2,3,5,6-tetraphenyl-
1,1,4,4-tetraiodo-1,4-disilacyclohexa-2,5-diene (9c)
1H-NMR (CDCl3): l 7.07–7.19 (m, 20H); 13C-NMR
(CDCl3): l 152.93, 137.50, 129.96, 127.32, 127.00. M.p.
(dec.): 200 °C. CIMS (NH3); m/z (relative intensity,
ion): 937.0, 0.67 [M+NH3]+. Crystals suitable for
X-ray diffraction were obtained by recrystallization
from methylene chloride.
2.1.14. Synthesis of 2,3,5,6-tetraphenyl-1,1,4,4-
tetrahydroxy-1,4-disilacyclohexa-2,5-diene 11
In a 50 ml flask equipped with a magnetic stirrer
were placed 9a (1.38 g, 2.50 mmol) and then acetone
(25 ml). After 9a had dissolved, distilled water (0.90 g,
50.0 mmol) was added dropwise through a syringe with
stirring whereupon a white precipitate was formed. The
reaction mixture was stirred at room temperature for 2
h and then filtered. The precipitate was washed with
3×5 ml of Et2O followed by 3×5 ml of pentane and
then the product was dried under reduced pressure.
2.1.11. Synthesis of 2,3,5,6-tetraphenyl-1,1,4,4-
tetramethoxy-1,4-disilacyclohexa-2,5-diene (9d)
In an NMR tube were placed 4 (55.4 mg, 0.100
mmol) and 1.2 ml of Et2O and then Si2(OMe)6 (48.4
mg, 0.20 mmol) was introduced through a syringe. The
mixture was vibrated for 5 h in an ultrasonic bath.
During the reaction, a white precipitate formed. The
volatile materials were removed under vacuum, and the
product was washed with 3×1 ml of pentane and then
dried under reduced pressure. Yield: 90%. 1H-NMR
(CDCl3): l 7.00–7.11 (m, 20H), 3.33 (s, 12H); 13C-
NMR (CDCl3): l 156.14, 140.82, 128.82, 127.60,
125.89, 50.96. M.p.: 188–190 °C. EIHRMS; Calc. for
C32H32O4Si2: 536.1839; found: 536.1841.
1
Yield: 98%. H-NMR (THF-d8): l 6.88–7.09 (m, 20H),
5.51 (s, 4H); 13C-NMR (THF-d8): l 157.21, 143.25,
130.20, 127.62, 125.51. M.p.: \300 °C. EIMS; m/z
(relative intensity, ion): 480.1 ([M+], 0.70); CIMS
(NH3, negative) 479.9 ([M−], 52.80). EIHRMS; Calc.
for C28H24O2Si2: 480.1213; found: 480.1221. Elemental
analysis: Calc. for C28H24O2Si2: C, 70.00; H, 5.00; Si,
11.67. Found: C, 70.57; H, 5.08; Si, 11.74%.
2.1.15. Synthesis of 1,4-dichloro-1,4-dimethyl-2,3,5,6-
tetraphenyl-1,4-disilacyclohexa-2,5-diene (12)
2.1.12. Synthesis of 2,3,5,6-tetraphenyl-1,1,4,4-
tetraphenoxy-1,4-disilacyclohexa-2,5-diene (9e)
The preparation of this compound was analogous to
that of 9a–c, except that 6 and SiCl4 were used as
starting materials. Although only one isomer of 6 (pre-
sumably the trans) was used in the reaction, a mixture
of diastereomers in a ca. 1:1 ratio was obtained. At-
tempts to separate these two isomers by fractional
sublimation or column chromatography failed. 1H-
NMR (CDCl3): l 6.98–7.13 (m, 20H), 0.37 (s, 3H),
0.30 (s, 3H); 13C-NMR (CDCl3): l 156.47, 156.44,
128.51, 128.38, 127.66, 127.60, 126.23, 0.40, 0.25. M.p.:
This synthesis is analogous to that of 9a–d, except
that P(OPh)3 was used as the derivatizing agent. The
product was washed with 3×1 ml pentane and dried
1
under reduced pressure. Yield: 95%. H-NMR (CDCl3):
l 6.68–7.12; 13C-NMR (CDCl3): l 157.24, 153.43,
139.16, 129.21, 128.76, 126.26, 122.07, 119.79. EIMS;
m/z (relative intensity, ion): 784.3 ([M+], 5.26).
EIHRMS; Calc. for C52H40O4Si2: 784.2465; found:

J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
281
\300 °C. EIHRMS; Calc. for C30H26Cl2Si2:
512.095015; found: 512.0959674.
centered in the X-ray beam using a video camera.
Crystal evaluation and data collection were performed
on a Bruker CCD-1000 diffractometer with a diffrac-
tometer-to-crystal distance of 5.08 cm. The initial cell
constants were obtained from three series of ꢀ scans at
different starting angles. Each series consisted of 20
frames at intervals of 0.3° in a 6° range about ꢀ with
an exposure time of 10 s per frame. The reflections were
successfully indexed by an automated indexing routine
built into the SMART program. The data were collected
using the hemisphere data collection routine. The recip-
rocal space was surveyed to the extent of 1.2 hemi-
2.1.16. Synthesis of trans-2,3,5,6-tetraphenyl-1,4-
dihydro-1,4-di-tert-butyl-1,4-disilacyclohexa-2,5-diene
(13)
The preparation of this compound is analogous to
that of 10 except that 8 and LiAlH4 were employed as
starting materials. The yield of 75%, includes about
10% of a minor (presumably cis) isomer. 1H-NMR
(CDCl3): l 6.82–7.07 (m, 24 H), 4.97 (s, 2 H, trans),
4.50 (s, 0.32 H, cis), 0.84 (s, 3.14 H, cis), 0.41 (s, 18 H,
trans); 13C-NMR (CDCl3) (trans): l 154.43, 143.59,
129.71, 127.49, 125.59, 28.34, 19.90; 13C-NMR (CDCl3)
(cis): l 156.69, 140.28, 129.34, 127.02, 125.43, 29.49,
20.00. M.p.: 238–240 °C. Crystals suitable for X-ray
diffraction spectroscopy were obtained by recrystalliza-
tion from methylene chloride.
,
spheres to a resolution of 0.80 A. The highly redundant
data sets were corrected for Lorentz and polarization
effects. The absorption correction was based on fitting
a function to the empirical transmission surface as
sampled by multiple equivalent measurements. A sum-
mary of crystallographic data including unit cells and
refinement data for each of the structure analyses ap-
pears in Table 1.
2.1.17. Variable temperature NMR studies of 8
The 1H- and 13C-NMR spectra of 8 were measured at
different temperatures to investigate restrictions on the
1
free rotation of the four phenyl groups. H-NMR at
3. Results and discussion
−30 °C (CDCl3): l 7.52–7.54 (d, 2H), 7.21 (t, 2H),
6.97 (t, 2H), 6.85 (t, 2H), 6.50–6.52 (d, 2H); 13C-NMR
(CDCl3): l 156.52, 141.55, 129.83, 129.64, 127.34,
126.90, 125.87, 26.47, 22.26. 1H-NMR at 25 °C
(CDCl3): l 6.52–7.54 (m, br, 20H), 0.51 (s, 18H);
13C-NMR (CDCl3): l 156.95, 141.79, 130.01, 127.18
3.1. Synthesis of 1,4-disilacyclohexa-2,5-dienes
In the reaction of diphenylacetylene with
Si2Cl(NMe2)5 (carried out in a sealed tube at 200 °C)
crystalline 4 was obtained in 45% yield, together with
the liquid by-product Si(NMe2)3Cl (reaction 3). The
constitution of the latter product was confirmed by
1H-NMR, 13C-NMR and MS spectroscopies. When
sym-Si2Cl2(NMe2)4was used in this reaction, however,
the isomeric products 5a and 5b were obtained in about
20% yield, accompanied by the formation of ca 5% of
the trimerization product hexaphenylbenzene. In this
case, one isomer, presumably trans-5, was the dominant
product, with a trace amount cis-5. The by-product, as
expected, was again Si(NMe2)3Cl (reaction 4).
1
(br), 125.99, 26.70, 22.27. H-NMR at 50 °C (CDCl3):
l 6.52–7.54 (br, 20H), 0.46 (s, 18H); 13C–NMR
(CDCl3): l 157.16, 141.92, 130.14, 127.20, 126.04,
26.80, 22.29. At temperatures below 0 °C, five different
protons were observed. At −30 °C, the 1H-NMR
spectrum also showed five well-separated peaks. More-
over, six distinct 13C signals were observed at that
temperature. From room temperature to 50 °C, two
pairs of 13C peaks associated with the phenyl groups
began to coalesce. Small changes in the chemical shifts
of the tert-butyl groups also occurred with temperature.
2.1.18. X-ray crystallographic determinations
A crystal of compound 4 was mounted on a glass
fiber on an Enraf–Nonius CAD4 for data collection at
293(2)91 K. The cell constants for the data collection
were determined from reflections found from a 360°
rotation photograph. Lorentz and polarization correc-
tions were applied. A nonlinear correction based on the
decay in the standard reflections was also applied to the
data. A series of azimuthal reflections was collected for
the crystal and a semi-empirical absorption correction
was applied to the data.
(3)
(4)
A silylene pathway similar to that proposed in the
literature [13] is suggested here as is illustrated in
Scheme 2 for the synthesis of 4. Amino groups (includ-
ing the dimethylamino group) are known to stabilize
silylenes [14], including an example (14) that is stable at
For compounds 8, 9c, 9e, 10 and 13, a crystal was
selected from oil under ambient conditions and was
attached to the tip of a glass capillary. The crystal was
mounted in a stream of cold nitrogen at 173(2) K and

282
J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
Table 1
Crystallographic data for structure refinement parameters for 4, 8, 9c, 9e, 10, and 13
4
8
9c
9e
10
13
Empirical formula C36H44N4Si2
C
38H38Cl2Si2
C29H21Cl3I4Si2
1039.59
173(2)
C52H40O4Si2
785.02
173(2)
C28H24Si2
416.65
173(2)
C36H40Si2
528.86
173(2)
Formula weight
Temperature (K)
588.93
293(2)
597.74
173(2)
0.71073
,
Wavelength (u, A) 0.71073
(Mo–Ka) 5025
0.71073
0.71073
0.71073
0.71073
Crystal system
Space group
Monoclinic
P21/n
Monoclinic
P21/c
Tetragonal
P4(2)nm
Monoclinic
P21/n
Monoclinic
P21/c
Triclinic
P1
Unit cell dimensions
,
a (A)
8.963(2)
15.510(3)
12.473(2)
90
98.15(3)
90
1716.4(6)
2
1.140
9.7708(8)
9.3984(8)
17.8150(14)
90
103.939(1)
90
1587.8(2)
2
1.250
11.7536(5)
11.7536(5)
12.2191(6)
90
90
90
1688.03(13)
2
2.045
9.1542(5)
17.1747(6)
15.6315(7)
90
97.649(1)
90
2010.26(17)
2
1.297
7.8373(4)
11.9759(7)
12.3495(7)
90
99.120(1)
90
1144.46(11)
2
1.209
12.0270(5)
69.5348(10)
12.1135(5)
69.5348(10)
77.0789(10)
63.6759(10)
1577.52(11)
2
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
V (A )
3
,
Z
Dcalc (Mg m−3
Absorption
coefficient
)
1.113
0.134
0.133
0.304
4.019
0.137
0.167
(mm−1
F(000)
)
632
632
972
824
440
568
Crystal size (mm)
Theta range for
data collection
(°)
0.4×0.4×0.3
2.11–27.47
0.46×0.40×0.10 0.35×0.35×0.30
0.45×0.45×0.40 0.45×0.40×0.35 0.50×0.40×0.40
2.36–26.37
2.40–28.30
1.95–26.37
2.38–27.37
1.69–26.37
Index ranges
−15h511,
−15k520,
−1651516
5025
−125h51,
105k511,
051522
9799
−155h515,
−155k515,
−1651515
14 104
−115h511,
05k517,
051519
17 677
−95h59,
05k514,
051515
8638
−145h515,
−135k515,
051516
Reflections
collected
13 896
Independent
reflections
Completeness to
theta=26.37°
(%)
3937 [Rint=0.0340] 3241
[Rint=0.0274]
99.8
2183 [Rint=0.0334] 4086
[Rint=0.0334]
99.5
2331
[Rint=0.0197]
99.8
6397
[Rint=0.0172]
99.2
NA
99.0
Absorption
correction
Refinement method Full-matrix
Semi-empirical
Semi-empirical
Full-matrix
Empirical with
SADABS
Empirical with
SADABS
Empirical with
SADABS
Empirical with
SADABS
Full-matrix
Full-matrix
Full-matrix
Full-matrix
least-squares on F2 least-squares on least-squares on F2 least-squares on least-squares on least-squares on F2
F2
F2
F2
Data/restraints/para NA
meters
32 41/0/184
2183/3/10
0.9474 and
0.9411
2331/0/144
6397/0/357
Final R indices
[I\2|(I)] a
R Indices (all
data) a
R1=0.0689,
R1=0.0333,
wR2=0.0782
R1=0.0530,
wR2=0.0835
1.031
R1=0.0275,
wR2=0.0775
R1=0.0286,
wR2=0.0782
1.052
R1=0.0338,
wR2=0.0854
R1=0.0451,
wR2=0.0907
1.035
R1=0.0314,
wR2=0.0841
R1=0.0385,
wR2=0.0879
1.048
R1=0.0442,
wR2=0.1092
R1=0.0494,
wR2=0.1092
1.038
wR2=0.1623
R1=0.1611,
wR2=0.2070
Goodness-of-fit on 1.147, 1.035
F2 b,c
Largest difference
0.306 and −2.58
0.328 and −2.07 1.920 and −0.744
0.325 and
0.326 and
0.367 and −0.292
peak and hole
−0.300
−0.200
−3
,
(e A
)
a R1=ꢀꢁFoꢂ−ꢂFcꢁ/ꢀꢂFoꢂ, wR2=[ꢀ[w(Fo2−F2c)2]/ꢀ[w(Fo2)2]]0.5, where w=1/[|2(Fo)2)+aP)2+bP+d+e sin q].
b GooF=[ꢀ[w(Fo2−Fc2)2]/(n−p)]0.5
.
c Twin component contributions were 0.52(3) and 0.48(3).
room temperature [15]. In an attempt to substantiate a
silylene pathway for the formation of 4 and 5a,b, we
carried out their syntheses in the presence of quinones
such as benzil, and dienes such as 1,3-butadiene which
are well known traps for R2Si: [16]. However, we
discovered that disilanes with electron-withdrawing
groups such as halogen and alkoxy reacted smoothly
with 1,2-quinones and 1,4-quinones to give doubly sily-

J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
283
lated products [17]. We therefore attempted to react
Si2(NMe2)5Cl or Si2(NMe2)4Cl2 with benzil in CHCl3
either at room temperature or at 75 °C. Because no
reaction was observed at either temperature, there was
a possibility that such a mixture at 180 °C (which is in
the temperature range where our reactions are carried
out) might give rise to silylene adducts that could be
trapped. When benzil and Si2(NMe2)5Cl were heated to
180 °C in a sealed tube, a brown insoluble material was
formed. Washing the reaction mixture with pentane
gave a gray precipitate, which appeared to be an
oligomeric material by 1H- and 13C-NMR
spectroscopies.
dered disilane, the generation of the silylene Si(NMe2)2
species via disproportionation of the disilane may be
more facile, owing to the less sterically hindered by-
product Si(NMe2)3Cl that is formed compared with
that formed in the Si2(NMe2)6/PhCꢁCPh, namely
Si(NMe2)4.
When the non-symmetric alkyne 1-phenyl-1-propyne
was allowed to react with Si2Cl(NMe2)5, the isomeric
products 7 were isolated (in ca. 10:1 ratio) in good
yield. Here trans-7 dominates according to 1H- and
13C-NMR spectroscopies (see Section 2). Similarly,
when sym-Si2Me2(NMe2)4 was allowed to react with
diphenylacetylene in a sealed tube, the corresponding
trans product 6 was obtained in 75% yield. Interest-
ingly, only a trace amount of cis-6 was detected. When
more bulky groups were introduced, much lower
product yields were obtained, even though higher tem-
peratures and prolonged reaction times were employed.
Thus in the reaction of sym-Sit2BuCl2(NMe2)2 with
diphenylacetylene, the corresponding trans product 8
was obtained exclusively in less than 5% yield at 210 °C
over a period of 3 days. In this reaction the silylene
SiCltBu is expected to be the reactive species.
Moreover, mass spectroscopy did not reveal the pres-
ence of
a silylene adduct. When trans,trans-1,4-
diphenyl-1,3-butadiene was used as a trap, however, an
HRMS of the reaction mixture gave a major peak for
15 (calc. for C20H26NSi: 322.18653; found: 322.186597)
thus supporting a silylene pathway for our reactions.
The varying yields obtained from diphenylacetylene
with the various dimethylamino disilanes can be at-
tributed to a combination of electronic effects on the
SiꢂSi bond strength and steric effects of the sub-
stituents. Thus the least sterically hindered disilane
Si2Me2(NMe2)4 gave product 6a (trans) in 75% yield,
whereas the highly bulky Sit2Bu2Cl2(NMe2)2 reacted in-
completely with diphenyl acetylene. It is apparent that
the presence of multiple dimethylamino groups on disi-
lanes favors disilane disproportionation reactions, per-
haps because the weakness of the SiꢂN bond (compared
with the SiꢂCl linkage [18]) facilitates the thermal cleav-
age of this bond, leading to the formation of reactive
silylene species that are electronically stabilized by one
or more NMe2 groups. As reported by us earlier [2],
when diphenylacetylene was allowed to react with disi-
lanes with zero, two, three, four, five and six dimethy-
lamino groups at 200 °C for 40 h, we found that the
ratio of disilacyclohexadiene to hexaphenylbenzene gen-
erally rose from zero in the case of the Si2Cl6/PhCꢁCPh
reaction (i.e. hexaphenylbenzene as the only detectable
product) to infinity (i.e., 9a as the sole detectable
When Si2(NMe2)6was allowed to react with dipheny-
lacetylene, not surprisingly, 4 was obtained. However,
the yield of this reaction dropped to ca. 20%, which is
less than half of that obtained from the Si2Cl(NMe2)5/
PhCꢁCPh reaction. This result may be rationalized on
the basis of the relative strengths of the SiꢂSi linkages
in the two disilanes, since a chloro substituent is ex-
pected to be more electron-withdrawing than
a
dimethylamino group, thereby weakening the SiꢂSi
bond. Moreover, Si2Cl(NMe2)5 is a less sterically hin-
product)
in
the
Si2(NMe2)5Cl/PhCꢁCPh
and
Si2(NMe2)6/PhCꢁCPh reactions. Because the Cl atom is
more electron-withdrawing than the NMe2 group [19],
the SiꢂSi link in Si2(NMe2)5Cl is weakened and the
SiꢂN bond is weaker than SiꢂCl bond, thus facilitating
disproportionation to give Si(NMe2)2 species. However,
as more NMe2 groups are substituted by Cl atoms, silyl
radical generation via homolytic SiꢂSi cleavage will be
increasingly favored over silylene production. Although
the mono-substituted disilane Si2(NMe2)Cl5 was not
obtained in pure form, the impure material led to
Scheme 2.

284
J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
Scheme 3.
3.2. Reactions of 1,4-disilacyclohexa-2,5-dienes 4, 6, 7,
8, 9a
hexaphenylbenzene as the only detectable product as
was the case with Si2Cl6.
Interestingly, silanes bearing dimethylamino groups,
such as Si(NMe2)3Cl and Si(NMe2)4 are unreactive
toward diphenylacetylene at 210 °C for 3 days. Prod-
ucts 4 and 5 might have been expected if silylene
formation had occurred. This observation further sup-
ports the idea that the presence of a SiꢂSi bond in
disilanes bearing dimethylamino groups favors the dis-
proportionation process involved in silylene generation.
Pyrolysis of either Si2(NMe2)5Cl or Si2(NMe2)4Cl2 at
200 °C in a sealed tube gave oligomeric products,
3.2.1. Halide substitution
It was found that the dimethylamino groups on 4, 6
and 7 are quite labile. When 4 was treated with excess
SiCl4 at 45 °C in Et2O for 3–4 h, 9a was obtained in
95% yield. Analogous treatment of 4 with PCl3 also
afforded 9a, and 9b was obtained analogously in the
reaction of PBr3 (or SiBr4) with 4. The reaction of 4
with PI3 gave product 9c in 70% yield.
Although trans-6 is formed in the Si2Me2(NMe2)4/
PhCꢁCPh reaction, treatment of this product with
SiCl4, gave both cis-12 and trans-12 in a ca. 1:1 ratio.
It is suggested that halogen substitution can occur via
an ionic pathway facilitated by the formation of a
five-coordinate silicon intermediate as shown in Scheme
3. A covalent pathway would seem to be more attrac-
tive, however, since the reaction occurs in a rather
nonpolar solvent (ether). The first two paths in Scheme
4 are expected to lead to stereoretention of each silicon
configuration in the substrate molecule, while the third
path in this scheme would result in the formation of the
opposite configuration at silicon. Apical departure of
the dimethylamino group in this path could be facili-
tated by a second SiCl4 molecule.
1
together with Si(NMe2)3Cl, as was confirmed by H-
and 13C-NMR, and mass spectroscopies. This is consis-
tent with our suggestion that in these disproportiona-
tion reactions Si(NMe2)2 [or Si(NMe2)Cl] together with
Si(NMe2)3Cl [or Si(NMe2)2Cl2] are formed. These
highly reactive silylenes could then lead to oligomeric
products. However, efforts to trap silylene species in
these experiments failed.
Interestingly, when phenylacetylene was allowed to
react with either Si2Cl(NMe2)5 or sym-Si2Cl2(NMe2)4,
no product analogous to 4 or 5 was detected. Instead,
the cyclodimerization product 16 was obtained as the
dominant product (85% conversion) together with the
two isomeric trimers 1,3,5-triphenylbenzene and 1,2,4-
triphenylbenzene (15% conversion) as shown by 1H-
and 13C-NMR, and GC-mass spectroscopies. This re-
sult is consistent with our earlier observations on the
tendency of terminally conjugated alkynes to undergo
thermal dimerization
3.2.2. Alkoxide and hydroxide substitution
When 4 was heated at 100 °C with excess Si2(OMe)6
in the absence of a solvent for 5–6 h, colorless crystals
of 9d formed in 80% yield. Compound 9e was obtained
as colorless crystals in 95% yield from the reaction of
P(OPh)3 and 4 at 140 °C for 15 h.
When water was added to a solution of 9a in ace-
tone, 11 formed as a white precipitate in 95% yield.
This product is very stable towards air and moisture,
and no decomposition was observed at 200 °C for 3
days. When 11 was treated with CaO at 250 °C under
vacuum in an attempt to dehydrate 11, the formation of
1
cis- and trans-stilbenes was detected by H- and 13C-
[2]. As expected, oligomeric material and Si(NMe2)3Cl
were also formed from the pyrolysis of these disilanes.
It should be noted that the triphenylbenzenes are also
formed in the absence of a disilane [2].
NMR spectroscopies. According to reaction 5, in which
the formation of a cyclic metasilicate such as Si3O69− is
suggested, the formation of stable volatile stilbene must

J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
285
constitute a substantial driving force for the cleavage of
the robust SiꢂO bond. However, a highly thermally
stable cross-linked polymer was obtained when the
dehydration reaction was carried out in refluxing acetic
anhydride.
bons to which they are bonded in each phenyl
substituent.
Variable temperature 1H- and 13C-NMR spectral
studies at 50 °C, ambient temperature and −30 °C
revealed changes consistent with restricted rotation of
the phenyl groups. Structure 17 represents a phenyl
group of 8 which is locked into an orientation wherein
its plane is perpendicular to that of the disilacyclohexa-
diene ring. The chemical environments of H1 and H1%
are different and the same is true of H2 and H2%, owing
to the trans orientations of the t-Bu and chloro groups
across the plane of the disilacyclohexadiene ring (see
Section 3.4). The VT 1H-NMR spectrum in Fig. 1a
reveals five chemical shifts (two doublets and three
‘triplets’) for the phenyl region of 8, plus a CHCl3
proton peak stemming from the CDCl3 solvent used.
We assign the two doublets to H1 and H1% since proton
couplings in phenyl rings generally decrease signifi-
cantly from ortho to meta to para oriented pairs of
hydrogens. Further support for these assignments
comes from the expectation that the difference in the
environments of these protons is maximized (Dl$1
ppm) by their proximity to the disilacyclohexadiene
ring. This reasoning suggests that the H2 and H2%
resonances should be overlapping doublets of doublets
with a smaller separation of their chemical shifts. Thus
we assign the resonances at 6.85 and 7.2 ppm to these
protons. H3 suffers no change in environment upon
rotation. Thus its multiplet resonance does not broaden
as significantly, and it persists even at room tempera-
ture (Fig. 1c).
(5)
3.2.3. Reduction of 9a with LiAlH4
When 9a was treated with LiAlH4 in refluxing THF,
the corresponding reduction product 10 was obtained
in 80% yield. Compound 10 was stable to purification
by column chromatography.
1
3.3. H- and 13C-NMR spectroscopic study
For compounds 4, 5, 6, 7, 9a–c, 10 and 11, the
1
aromatic H-NMR peaks all appeared below 7.24 ppm,
the chemical shift for CHCl3 in the NMR solvent
CDCl3. For compound 8 (which has the trans configu-
ration), however, aromatic chemical shifts as high as
7.54 ppm were also evident. In fact, four equally spaced
broad singlets (l 7.54, 7.18, 6.85, 6.52 ppm) accompa-
nied by a triplet at 6.96 ppm were observed. Similarly,
the 13C-NMR spectrum displayed four singlets (l
156.95, 141.79, 130.01, 125.99, the latter two somewhat
broadened) and a broad peak at 127.18 ppm. Because
the tert-butyl group in 8 is the most sterically hindered
group in this series of compounds, this observation
could be attributed to restricted rotation of the four
phenyl groups owing to the steric encumbrance of the
tert-butyl groups. Such restricted rotation would be
expected to differentiate the five protons and the car-
Scheme 4.

286
J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
ment is unequivocally indicated for reasons that are not
clear.
3.4. Structural considerations
Molecular structure determinations by X-ray crystal-
lography were carried out on 4, 8, 9c, 9e, 10 and 13. All
these compounds featured a planar 1,4-disilacyclohexa-
2,5-diene six-membered ring, with four phenyl groups
oriented nearly perpendicular to the central ring. Be-
cause all of these structures possess an inversion center
in the middle of the central ring, these compounds
possess C2 symmetry when all four silicon substituents
are identical, except in 4 and 9e wherein the silicon and
ring carbon phenyls are not crystallographically equiva-
lent. Although 10 could possibly have higher symmetry,
close inspection indicates that there are also two sets of
crystallographically distinguishable phenyl groups in
this compound, therefore lowering the expected D2h
symmetry.
The structures for these compounds are shown in
Figs. 3–8 with selected bond distances and angles for
each. Compounds 8 and 10 represent the two extremes
of this series of 1,4-disilacyclohexa-2,5-dienes in terms
of relative steric encumbrance. However, the bond dis-
tances and bond angles of the central ring of these two
structures are within three times their ESD values. For
example, the average SiꢂC bond distance in 8 is
Fig. 1. 1H-NMR spectra for 8 at: (a) −30; (b) 20; and (c) 50 °C.
,
,
1.8798(17) A the CꢀC bond distance is 1.362(2) A.
,
,
These distances are 1.8709(13) A and 1.3550(18) A,
respectively, in 10. The bond angles on the central ring
of 8 and 10 are CꢂSiꢂC% 111.67(7), SiꢂC%ꢂC¦, 125.25(12)
and CꢂSiꢂC% 113.38(6), SiꢂC%ꢂC¦ 124.02(10), respec-
tively. The differences between other corresponding
angles are even smaller.
Fig. 2. 13C-NMR spectra for 8 at: (a) −30; (b) 20; and (c) 50 °C.
The VT 13C-NMR spectra (Fig. 2) show the ipso-car-
bon at ca. 141 ppm as expected, and the para-carbon at
about 126 ppm. Also as expected, the ortho- and meta-
carbon pairs coalesce upon warming. It is tempting to
assign chemical shifts to these pairs which are centered
at ca. 127 ppm for C1, C1% and at about 130 ppm for
C2, C2% based on their relative separation (as was done
for H1, H1% and H2, H2%). However, the 2D COSY
spectrum does not bear this out and the reverse assign-
Fig. 3. Computer drawing of the molecular structure of 4. Selected
,
bond distances (A) and bond angles (°): C(1)ꢂSi(1): 1.878(3),
C(2)ꢂSi(1): 1.881(3), Si(1)ꢂN(1): 1.703(3), Si(1)ꢂN(2): 1.707(3),
C(1)A1ꢂC(2)ꢂSi(1):
125.3(2),
C(1)A1ꢂC(2)ꢂC(9):
122.1(3),
N(1)ꢂSi(1)ꢂN(2): 108.4(2), C(1)ꢂSi(1)ꢂC(2): 110.35.

J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
287
Attempts to solve the molecular structure of 12 were
not completely successful owing to disorder of the
methyl and chloro substituents. Thus although the
configuration of the rings could be ascertained, the
present isomer could not. Crystals of 6 suffered from
severe disorder problems. As a result, no meaningful
information regarding the isomer present could be ob-
tained. It is likely on steric grounds, however, that the
Me2N groups are trans to each other.
Fig. 4. Computer drawing of the molecular structure of 8. Selected
,
bond distances (A) and bond angles (°): C(5)ꢂSi(1): 1.8824(17),
C(5)ꢂC(6): 1.362(2), C(1)ꢂSi(1): 1.8971(17), C(5)ꢂC(7): 1.505(2),
Si(1)ꢂCl(1): 2.0853(6), C(5)ꢂSi(1)ꢂC(6A): 111.67(7), Si(1)ꢂC(5)ꢂC(6):
125.25(12), Cl(1)ꢂSi(1)ꢂC(1): 107.07(5), C(5)ꢂC(6)ꢂSi(1A) 123.06.
Fig. 6. Computer drawing of the molecular structure of 9e. Selected
,
bond distances (A) and bond angles (°): Si(1)ꢂC(13): 1.8661(14),
Si(1)ꢂO(1): 1.6287(10), Si(1)ꢂO(2) 1.6334(10), Si(1)ꢂC(20): 1.8569(14),
C(13)ꢂC(20A):
1.350(2),
C(13)ꢂSi(1)ꢂC(20):
113.41(6),
Si(1)ꢂC(13)ꢂC(20A): 123.36(11), C(13A)ꢂC(20)ꢂSi(1): 123.16(11),
O(1)ꢂSi(1)ꢂO(2): 103.88(5).
Fig. 5. Computer drawing of the molecular structure of 9c. Selected
,
bond distances (A) and bond angles (°): Si(1)ꢂC(1): 1.862(3),
Si(1)ꢂC(1A): 1.862(3), Si(1)ꢂI(1) 2.4339(19), Si(1)ꢂI(2): 2.4382(18),
C(1)ꢂC(1A): 1.362(6), C(1)ꢂSi(1)ꢂC(1A): 115.3(2), Si(1)ꢂC(1)ꢂC(1A):
122.33(10), I(1)ꢂSi(1)ꢂI(1A): 108.61(5).
In compound 9e there are eight phenyl groups ren-
dering the molecular quite bulky. Although the central
part is essentially the same as in other compounds in
this series, the behavior of the two phenoxy groups is
interesting. A side view of the structure indicates that
there are two sets of phenoxy groups, with two phenyl
groups in each set. One set is almost in parallel to the
central 1,4-disilacyclohexa-2,5-diene ring while the
other is close to being perpendicular to it. This is
caused either by the steric hindrance of the phenoxy
groups, or electronic repulsion effects from the ring
aromatic pi systems.
Fig. 7. Computer drawing of the molecular structure of 10. Selected
,
bond distances (A) and bond angles (°): C(1)ꢂSi(1): 1.8695(13),
Si(1)ꢂC(2A): 1.8723(13), C(1)ꢂC(2): 1.3550(18), C(1)ꢂC(3):
1.4978(17), C(1)ꢂSi(1)ꢂC(2A): 113.38(6), Si(1)ꢂC(1)ꢂC(2): 124.02(10),
C(1)ꢂC(2)ꢂSi(1A): 122.58(10).

288
J. Yang et al. / Journal of Organometallic Chemistry 649 (2002) 276–288
References
[1] J. Yang, J. Verkade, J. Am. Chem. Soc. 120 (1998) 6834.
[2] J. Yang, J. Verkade, J. Organometallics 19 (2000) 893.
[3] E. Chernyshev, N. Komalenkova, S. Bashkirova, USSR. Zh.
Obshch. Khim. 41 (1971) 1175.
[4] (a) W. Atwell, D. Weyenberg, J. Am. Chem. Soc. 90 (1968) 3438;
(b) E. Chernyshev, N. Komalenkova, S. Bashkirova, J. Gen.
Chem. USSR 41 (1971) 1177;
(c) E. Chernyshev, N. Komalenkova, S. Bashkirova, V. Sokolov,
J. Gen. Chem. USSR 48 (1978) 757;
(d) K. Welsh, J. Corey, Organometallics 6 (1987) 1393;
(e) G. Maiser, H. Reisenauer, K. Schottler, U. Wessolek-Kraus,
J. Organomet. Chem. 366 (1989) 25;
(f) G. Maiser, K. Schottler, H. Reisenauer, Tetrahedron Lett. 26
(1985) 4079.
[5] Y. Tanaka, H. Yamashita, M. Tanaka, Organometallics 14 (1995)
530.
[6] A. Medvedeva, G. Lyashenko, O. Kozyreva, M. Voronkov, Russ.
J. Gen. Chem. 65 (1995) 145.
[7] H. Sakurai, T. Koyama, M. Kira, A. Hosomi, Y. Nakadaira,
Tetrahedron Lett. 23 (1982) 543.
[8] A. Naka, K. Yoshizawa, S. Kang, T. Yamabe, M. Ishikawa,
Organometallics 17 (1998) 5830.
[9] (a) E. Wiberg, O. Stecher, A. Neumaier, Inorg. Nucl. Chem. Lett.
1 (1965) 33;
Fig. 8. Computer drawing of the molecular structure of 13. Selected
,
bond distances (A) and bond angles (°): C(5)ꢂSi(1): 1.8822(16),
Si(1)ꢂC(12A): 1.8842(16), C(5)ꢂC(12): 1.356(2), C(5)ꢂC(6): 1.500(2),
C(5)ꢂSi(1)ꢂC(12A):
111.70(7),
Si(1)ꢂC(5)ꢂC(12):
123.74(12),
C(5)ꢂC(12)ꢂSi(1A): 123.90(12).
(b) Y. Wan, J. Verkade, Inorg. Chem. 32 (1993) 431.
[10] E. Hengge, H. Pletka, F. Hoefler, Monatsh. Chem. 101 (1970) 325.
[11] B. Reiter, K. Hassler, J. Organomet. Chem. 467 (1994) 21.
[12] Y. Wan, PhD Thesis, Iowa State University, 1994.
[13] (a) T. Barton, J. Kilgour, J. Am. Chem. Soc. 98 (1976) 7746;
(b) E. Chernyshev, N. Komalenkova, S. Bashkirova, V. Sokolov,
USSR. Zh. Obshch. Khim. 48 (1978) 830;
(c) E. Chernyshev, N. Komalenkova, S. Bashkirova, USSR. Dokl.
Akad. Nauk SSSR 205 (1972) 868.
[14] B. Gehrhus, M. Lappert, Polyhedron 17 (1998) 999 (and references
therein).
[15] M. Denk, R. Lennon, R. Hayashi, R. West, A. Belyakov, H.
Verne, A. Haaland, M. Wagner, N. Metzler, J. Am. Chem. Soc.
116 (1994) 2691.
[16] (a) S. Tsutsui, K. Sakamoto, M. Kira, J. Am. Chem. Soc. 120
(1998) 9955;
4. Supplementary material
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC nos. 1755311, 175304, 175306,
175307, 175305, 75303 and 175308 for compounds 4, 8,
9c, 9e, 10, 12 and 13. Copies of this information may be
obtained free of charge from The Director, CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (Fax: +44-
1223-336033; e-mail: deposit@ccdc.cam.ac.uk or www:
http://www.ccdc.cam.ac.uk).
(b) Y. Kwak, K. Lee, J. Organomet. Chem. 542 (1997) 219;
(c) D. Lee, C. Kim, M. Lee, Phosphorus, Sulfur Silicon Relat.
Elem. 119 (1996) 37 (and references therein).
[17] J. Yang, J. Verkade, J. Organomet. Chem., accepted for publica-
tion.
Acknowledgements
[18] T. Moeller, Inorganic chemistry, Wiley, New York, 1982, 238pp.
[19] The electronegativity of an NMe2 group is expected to be lower
than that of a Cl atom on the Pauling electronegativity scale. See:
J.E. Huheey, E.A. Keiter, R.L. Keiter, Inorganic Chemistry,
Fourth Edition, Harper Collins College Publishers, New York,
1993, 197pp.
The authors thank the National Science Foundation
for grant support, Dr Philip Kisanga for helpful sugges-
tions and Professors Andrzej Wroblewski and Robson
Matos for assistance.