Magnesium and chlorostannanes - Building blocks for novel tinmodified silanes

Article abstract of DOI:10.1016/j.jorganchem.2003.07.001

Cyclic stannasilanes with alternating Si-Sn sequences ([Ph₂ Sn-Me₂Si]₃, 1; [t-Bu₂Sn-Me₂ Si]₂, 2) were synthesized by the reaction of dimethyldichlorosilane with diorganodichlorostannanes (R

Full text of DOI:10.1016/j.jorganchem.2003.07.001

Journal of Organometallic Chemistry 686 (2003) 332Á340
/
www.elsevier.com/locate/jorganchem
Magnesium and chlorostannanes*
/
building blocks for novel
tinmodified silanes
P. Bleckmann c, T. Bruggemann c, S.V. Maslennikov b,*, T. Schollmeier c,
¨
M. Schurmann a, I.V. Spirina b, M.V. Tsarev b, F. Uhlig a,*
¨
a Institute for Inorganic Chemistry, Graz University of Technology, Stremayrgasse 16, A-8010 Graz, Austria
b Department of Chemistry, Nizhni Novgorod State University, Gagarin av. 23/5, Nizhni Novgorod 603950, Russia
c Fachbereich Chemie der Universitat Dortmund, Otto-Hahn-Straße 6, D-44221 Dortmund, Germany
¨
Received 16 May 2003; received in revised form 28 July 2003; accepted 28 July 2003
Dedicated to Prof. Dr. H. Marsmann on the occasion of his 65th birthday
Abstract
Cyclic stannasilanes with alternating SiÁ
of dimethyldichlorosilane with diorganodichlorostannanes (R2SnCl2, Rꢀ
pathway towards 2 via the unexpected intermediate product ClÁSiMe2Á(t-Bu2Sn)2Á
/
Sn sequences ([Ph2SnÁ
/
Me2Si]3, 1; [t-Bu2SnÁ
Ph, t-Bu) in the presence of magnesium. The reaction
SiMe2ÁCl (5) is discussed. In addition, some
/
Me2Si]2, 2) were synthesized by the reaction
/
/
/
/
/
thermodynamic and kinetic studies for the formation of the rings (1, 2) and the Grignard-type tin compounds are presented. All
compounds were characterized by NMR, MS and elemental analysis. The solid state structures of 1, 2 and 5 were determined by X-
ray crystallography.
# 2003 Elsevier B.V. All rights reserved.
Keywords: Cyclic stannasilanes; Magnesium; X-ray crystallography
1. Introduction
where Xꢀ
R5ꢀ
halogen, alkyl, aryl.
Depending on the Si-chain length of the used a,v-
dihalodimethylsilanes (XÁ(SiMe2)n ÁX; XꢀF, Cl; nꢀ
2Á6) a large variety of novel monocyclic SiÁSn rings had
been obtained. In continuation of this work, we
reported also the synthesis of a tin and silicon contain-
ing bicyclo[2.2.2]octane (compound F). However, only
/
Cl, F; R1Á
/
R4ꢀ
/
alkyl, aryl, silyl, stannyl; R,
/
Although, the chemistry of mono and polycyclic
silanes and siloxanes has been well investigated during
the last decades, only little is known about cyclic silanes
containing elements of Groups 13, 14, or 15. In the last
few years we described the syntheses of open-chain and
cyclic tin-modified silanes in simple one pot synthesis by
reacting dimethyl- or diphenyldichlorostannanes with
diorganodichloro- or -difluorosilanes in the presence of
/
/
/
/
/
/
ring sizes ranging from 6 to 8 could be obtained (A, nꢀ
/
2; B, nꢀ3; C, nꢀ4; D, nꢀ5; E, nꢀ6; F, nꢀ4) if
/
/
/
/
/
magnesium and THF as solvent [1Á
/4].
diphenyl- or dimethyldichlorosilanes are used as starting
materials (Chart 1).
This paper is dealing with the reactions of the
dimethyldichlorosilane the missing silane in the row of
the Si-containing starting materials (see Chart 1) with
different diorganodichlorostannanes in the presence of
magnesium. The synthetic and structural investigations
of the reactions and reaction pathways are accompanied
by kinetic studies and theoretical calculations.
ꢂMg
0
RR1R2SiꢁXꢂClꢁSnR3R4R5
ꢁSnR3R4R5
RR1R2Si
ꢁMg(X)Cl
(1)
* Corresponding author.
0022-328X/03/$ - see front matter # 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2003.07.001

P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á
/
340
333
Chart 1. Reaction products of XÁ
/
(SiMe2)n Á
/
X (Xꢀ
/
F, Cl; nꢀ
/
2Á/6)
with R2 SnCl2 in the presence of magnesium (A, nꢀ
/
2; B, nꢀ3; C,
/
?
nꢀ
/
4; D, nꢀ
/
5; E, nꢀ
/
6; F, nꢀ4).
/
Chart 2. Heat of formation (DHf) for compounds 1, 2 and the
hypothetical rings 4 and 3. Reaction enthalpy (DHr) for the conversion
of 1 into 4 and 2 into derivative 3.
2. Results and discussion
2.1. Syntheses of novel tin-modified silanes
in the reaction ethalpies of 163.2 kJ molꢁ1 indicates a
higher stability of the four-membered t-butyl substi-
tuted ring 2 in comparison to the hypothetical phenyl
substituted derivative 4 and display from a thermody-
namical point of view the different reaction behavior of
Ph2SnCl2 or t-Bu2SnCl2.
The reaction of equimolar amounts of dimethyldi-
chlorosilane (Me2SiCl2) and diphenyldichlorostannane
(Ph2SnCl2) with magnesium provided the 1,3,6-trisila-
2,4,6-tristannacyclohexane (1) in nearly quantitative
yield (Scheme 1). The four-membered 1,3-disila-2,4-
distannacyclobutane (2) was observed by reacting Me2-
SiCl2 with the sterically more demanding di-tert-butyl-
dichlorostannane in yields between 80 and 95%.
The different reaction behavior of t-Bu2SnCl2 and
Ph2SnCl2 can be explained by the different steric
demand of the substituents at the tin atoms as well as
by some thermodynamic considerations.
The stability of the synthesized derivatives 1 and 2 as
well as of the hypothetical compounds 3 and 4 was
estimated by theoretical calculations (see Section 3.5).
The calculated heats of formation are given in Chart 2.
For a conversion of the six-membered ring 1 into the
Surprisingly, one of the first steps in the reaction
pathway towards the ring compound 2 is the formation
of the distannane 5 (ClSiMe2Á
/
t-Bu2SnÁ
/
t-Bu2SnÁSi-
/
Me2Cl). This result was proven by time dependent
29Si- and 119Sn-NMR spectra (for 119Sn-NMR see Fig.
1). The 119Sn-NMR signal of the starting material at
ꢀ
/
55 ppm (spectra A) disappeared within 1 h yielding
one new singlet at ꢁ122.5 ppm for ClSiMe2Át-Bu2SnÁ
t-Bu2SnÁSiMe2Cl (5, spectra B). At this stage the
/
/
/
/
reaction can be stopped by quickly replacing the original
solvent THF with n-hexane and we were able to isolate
and characterize compound 5 by NMR and X-ray
crystallography. If the solvent was not changed two
four-membered ring 4 a reaction enthalpie of ꢁ209.3 kJ
/
new signals were observed after 3 h at ꢁ
18.2 ppm, and ꢂ10.8 ppm (spectra C). The share of
the two major signal at ꢁ122.5, and ꢁ40.2 ppm
changed slowly within the next 22 h (spectra D). After
24Á25 h only one major signal at ꢁ40.2 ppm (2) was
remaining (spectra E), suggesting that the unidentified
compounds with 119Sn-NMR shifts at ꢂ
10.8 and
18.2 ppm were intermediate products.
/40.2 ppm (2),
molꢁ1 was found. The enthalpy of reaction for the
transformation of compound 3 into derivative 2 was
ꢁ
/
/
/
/
calculated with ꢁ
372.5 kJ molꢁ1. Although, these
considerations are gas phase calculations, the difference
/
ꢀ
/
/
/
/
ꢁ
/
An identical time dependency was achieved by a series
of 29Si-NMR experiments. Although, there is only little
experimental support for the conversion of compound 5
into the distannane 2, we suggest that an insertion of
magnesium into the tinÁtin bond of derivative 5 is the
/
key step resulting in the magnesium stannide 6 (Scheme
2). The ring closure can occur by ionic mechanism or
alternatively by the intermediate formation of a sila-
stannene (Me2SiÄ
/
SnÁt-Bu2), however, the experimental
/
data do not allow to decide between the two possible
reaction pathways.
?
Scheme 1. Reaction of Me2SiCl2 with R2SnCl2 (R?ꢀ
Mg.
/
t-Bu, Ph) and

334
P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á340
/
Fig. 1. Time-dependent 119Sn-NMR spectra for the reaction of t-Bu2SnCl2, Me2SiCl2 and Mg. (119Sn{1H}-NMR: 111.91 MHz; solvent THF, D2O-
cap.; spectra A: 24 scans; spectra BÁE: 1024 scans; room temperature.)
/
materials, traces of oxygen or moisture and the con-
centration. Larger amounts of the solvent (THF) yielded
an extended reaction time and therefore a prolonged
time for the isolation of compound 5. Although, only
few examples are known in literature [7Á13] we suspect
/
that in all cases of interaction of organotin halides and
organosilicon halides in the presence of magnesium, the
initial step of the reactions is the formation of Grignard
?
analogue tin compounds of type R2Sn(MgÁCl)Cl or
/
(R?R2Sn)MgÁ
/
X (Xꢀhalogen, SnR2R?). Therefore, we
/
tried to get an independent prove for the formation of
such a Grignard-type derivative. Due to the complex
reaction pathway for the synthesis of compound 2, we
changed to a simpler system (Ph2SnCl2Á
/
Ph3SiXÁMg)
/
for these investigations.
Scheme 2. Possible reaction mechanism for the formation of com-
pound 2.
2.2. Formation of SnÁMg bonds containing intermediates
/
Insertion reactions of alkaline earth metals into tinÁ
/
tin bonds are well known from our own work as well as
from literature (Eq. (2)) [5,6].
The interaction between Ph2SnCl2 and Ph3SiF in the
presence of Mg yielded the octaphenyl-1,3-disila-2-
ꢂMII
stannapropane (Ph3SiÁ
was used as a model system for studying regularities
and mechanism of the formation of SiÁSn bonds.
According to literature [8Á10] final products of dialkyl-
and diarylstannane dichlorides reduction by magnesium
/
SnPh2Á
/
SiPh3). This reaction
R?R2SnꢁSnR2R? 0 R?R2SnꢁMII ꢁSnR2R?
(2)
where MIIꢀ
It should be carefully noted that the reaction time is
highly sensitive towards small impurities of the starting
/alkaline earth metal (Mg, Ca, Sr, Ba)
/
/

P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á
/
340
335
are cyclic 5Á
whereas the presence of triphenylfluorosilane in the
reaction leads to Ph3SiÁSnPh2ÁSiPh3 as major product
(Eq. (3)).
/7-membered polystannanes (R2Sn)5Á7,
/
/
R2SnCl2 ꢂ2Ph3SiFꢂ2Mg 0 Ph3SiÁSnR2SiPh3
ꢂ2MgClF
(3)
The formation of the Grignard analogue compounds
?
?
R2Sn(MgÁ
key step of this reaction. In order to prove the presence
of species containing a SnÁMg bond probes were taken
/
Cl)Cl and R2SnSiPh3(MgÁ
/
Cl) should be the
/
from the reaction mixture of Ph2SnCl2 with magnesium
in THF at different reaction times. The solution was
separated from the magnesium metal and then reacted
with 1,2-dibromoethane. According to the reaction (Eq.
(4)) [7] the amount of the ethylene evolved is equal to the
Fig. 3. The dependence of magnesium oxidation rate on THF
concentration, mixture THFÁtoluene, C (Ph2SnCl2)ꢀ
0.2 mol lꢁ1
", 293 K; j, 303 K; ', 313 K
/
/
.
number of SnÁMg bonds in the solution worked up with
/
1,2-dibromoethane.
R3SnMgClꢂC2H4Br20 R3SnBrꢂMgClBrꢂC2H4
The rate of magnesium oxidation by Ph2SnCl2 in the
toluene has a maximum at the oxidizer’s
concentration of 0.2 mol lꢁ1. Further increasing of the
concentration of Ph2SnCl2 leads to a decreasing of the
reaction rate (Fig. 2). The rate of the reaction vs. the
mixture THFÁ
/
(4)
The concentration of the SnÁ
/
Mg bonds (for
cPh SnCl /293 K) was found with
ꢀ
/
0.25 M lꢁ1 and Tꢀ
0.008 mol per mol of the starting material Ph2SnCl2
after 40 min, 0.09 after 160 min and only traces after 400
2
2
concentration of THF in the mixture THFÁtoluene with
/
diphenyltin dichloride concentration remaining constant
is depicted at the Fig. 3.
min. The time dependence of the concentration of SnÁ
/
It has been shown in literature [14] that the mechan-
ism of a Grignard reagent formation is quite complex.
On the other hand the kinetics of many reactions of
metals with organic oxidizing agents (alkyl halides,
metal chlorides containing organic ligands, etc.) in
non-aqueous solutions can be adequately described for
Mg bonds confirms their intermediate formation in the
reaction.
An additional prove of the proposed formation of
SnÁMg bonds was obtained by kinetic measurements.
/
The dependence of the reaction rate on the Ph2SnCl2
concentration in THF in the absence of fluorosilane is
linear within the concentration range of Ph2SnCl2 from
0 to 0.3 mol lꢁ1. Since 0.3 mol lꢁ1 is the maximal
example by using the LangmuireÁ/Hinshelwood or the
EalyÁRedeal mechanism [15]. However, dependencies
/
obtained for the oxidation of magnesium by diphenyltin
dichloride do not correspond to any of those schemes.
solubility of Ph2SnCl2 in THF, the mixture THFÁ
toluene (CTHF
6.9 mol lꢁ1) was used to extend the
dependence to higher concentrations of the oxidizer.
/
ꢀ
/
The closest approximation is the LangmuireÁHinshel-
/
Fig. 2. The dependence of magnesium oxidation rate on Ph2SnCl2 concentration, mixture THFÁ
/
toluene, C (THF)ꢀ
6.9 mol lꢁ1. (1) 293 K; (2) 303
/
K; (3) 313 K; (4) theoretical, in accordance with Eq. (8), calculated for the dependency at 303 K.

336
P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á340
/
wood mechanism with the adsorption of reagents on the
same reaction center.
KR1
R1ꢂS0 X R1S
KR2
(5)
(6)
(7)
R2LꢂS0 X R2S
k
R1SꢂR2S 0 products
R1 and R2 are the reagents participating in the reaction,
KR1 and KR2 are their equilibrium constants of adsorp-
tion on the metallic surface, S0 is an active center of the
metallic surface, k is the surface rate constant.
The rate of reaction is described by the expression
k+KR1+KR2+CR1+CR2
(1 ꢂ KR1+CR1 ꢂ KR2+CR2)2
V ꢀ
(8)
where CR1 and CR2 are reagent’s concentration.
Theoretical curve obtained in accordance with Eq. (8)
for the dependence at 303 K (with the adsorption
equilibrium constants 112.5 for Ph2SnCl2 and 1.9 for
Fig. 4. View of a molecule of 1 showing 30% probability displacement
ellipsoids and the atom numbering. Hydrogen atoms bonded to carbon
have been removed for clarity.
THF and surface reaction rate constant 3.6ꢃ g
10ꢁ5
/
cmꢁ2 minꢁ1) shows the same tendency but does not
repeat the experimental curve exactly (Fig. 2), especially
at higher concentrations of the oxidizer. As shown
above, the products of the interaction between Ph2SnCl2
and magnesium in the mixture THFÁtoluene might
/
contain species with one or more tin-tin bonds that may
be able to adsorb on the magnesium surface. It is
possible that at higher Ph2SnCl2 concentrations such
compounds with tinÁtin bonds are formed, causing the
/
discrepancies between theoretical and experimental rates
of reaction. However, no experimental proof for this
fact was given from the investigation of the reaction
mixture.
The rate of magnesium oxidation does not depend on
the amount of Ph3SiF within the concentration range 0Á
0.4 mol lꢁ1. Thus, the synthesis of compounds contain-
ing bonds SiÁSn is most likely accomplished via the
formation of transient organometallic species containing
the MgÁSn bond. Similar results can be observed by
/
/
Fig. 5. View of a molecule of 2 showing 30% probability displacement
ellipsoids and the atom numbering. Hydrogen atoms bonded to carbon
have been removed for clarity.
/
using Ph3SiCl instead of the fluoro derivative.
Ph2Sn(MgCl)Cl has not been obtained as a pure
compound yet. The closest analog is bis-(triphenyltin)
magnesium that was observed by the reaction of
magnesium with triphenyltin chloride in THF [14,16].
This compound is unstable even in the absence of
oxygen and moisture and decomposes in solution by
reacting with excess magnesium or THF [16].
Ph2Sn(MgCl)Cl should be even more unstable.
details, and selected interatomic parameters are sum-
marized in Tables 1 and 2.
2.3.1. Molecular structure of 1
Compound 1, depicted in Fig. 4, shows a six-
membered ring comprised of an alternating SiÁSn
/
sequence. The ring conformation is that of a classical
chair resulting from the approximate tetrahedral geo-
metry of the ring atoms. Slight distorsions from ideal
tetrahedral geometry are due to the different steric
demand of the ligands at the silicon and tin centers.
2.3. Structural discussion of 1, 2 and 5
Compounds 1, 2 and 5 are crystalline solids and we
were able to isolate these derivatives by crystallization
from n-hexane. The molecular structures of 1, 2 and 5
SiÁ
and 115.1(9)8 while SnÁ
109.7(6) up to 110.7(5)8. SiliconÁ
/
SnÁ
/
Si bond angles are observed between 113.9(5)
SiÁSn bond angles reach from
tin contacts are ob-
/
/
are presented in Figs. 4Á/6. Unit cell data, refinement
/

P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á
/
340
337
backbone of 2 displays a nearly square planar four-
membered ring with bond angles of 90.19(4)8 for Si(1)Á
Sn(1)ÁSi(1A) and 89.81(4)8 for Sn(1)ÁSi(1)ÁSn(1A).
The tinÁSilicon distances are observed at 2.610(2) and
/
/
/
/
/
˚
2.620(1) A and therefore in ranges found in the related,
less strained six-membered ring [Át-Bu2SnÁSiMe2Á
2.597(3) A) as well as the five-
membered ring Ph2Si[t-Bu2SnÁSiMe2]2 [17] (2.597(3)Á
carbon distances are found at
/
/
/
˚
SiMe2Á
/
]2 [17] (2.586(3)Á
/
/
/
˚
2.623(3) A). TinÁ
/
˚
2.207(4) and 2.208(5) A while siliconÁ
/
carbon distances
˚
are observed between 1.886(4) and 1.892(4) A.
2.3.3. Molecular structure of 5
The distannane derivative 5 is depicted in Fig. 6. The
disordered tert-butyl und chlorodimethylsilyl groups for
compound 5 were found with occupancies of 0.2 (Si(2?),
C(2?)), 0.3 (Cl(1?), Cl(2?), C(1?), C(2?)), 0.7 (Cl(1), Cl(2),
C(1), C(2)) and 0.8 0.2 (Si(2), C(2)). One C atom and the
Cl atom of a chlorodimethylsilyl group are disordered
over two positions sharing their positions not exactly.
Two pairs (Cl(1), C(1?)) and (Cl(1?), C(1)) have been
refined with the same anisotropic displacement para-
meters (EADP [18]). At the other chlorodimethylsilyl
group one C atom and the Cl atom are sharing their
positions exactly (EXYZ [18]). Two pairs (Cl(2), C(2?))
and (Cl(2?), C(2)) have been refined with the same
anisotropic displacement parameters (EADP [18]). Such
a disorder of chloro and methyl substituents were found
Fig. 6. View of a molecule of 5 showing 30% probability displacement
ellipsoids and the atom numbering. Hydrogen atoms bonded to carbon
have been removed for clarity.
˚
served between 2.566(2) and 2.577(2) A, and therefore in
ranges observed previously in related compounds [3,4].
2.3.2. Molecular structure of 2
Compound 2, shown in Fig. 5, crystallizes in the
¯
triclinic space group P1. Due to the, in contrast to 1
higher ring strain as well as the bulkiness of the sterically
more demanding tert-butyl groups the central SiÁSn
/
Table 1
Crystal data and structure refinement for 1, 2 and 5
1
2
5
Formula
fw (g molꢁ1
Crystal system
C42 H48 Si3 Sn3
993.14
Triclinic
C20 H48 Si2 Sn2
582.14
Triclinic
C20 H48Cl2 Si2 Sn2
653.04
Monoclinic
)
Crystal size (mm3)
Space group
0.08ꢃ
/
0.08ꢃ/0.06
0.1ꢃ
/
0.08ꢃ/0.08
0.18ꢃ
/
0.15ꢃ0.15
/
¯
P1
¯
P1
P21/c
Unit cell dimensions
˚
a (A)
˚
10.8740(3)
11.9251(3)
18.4995(6)
73.5830(12)
81.5325(12)
72.7784(13)
2192.63(11)
2
8.8048(7)
8.9379(6)
9.7092(9)
78.415(5)
82.359(4)
65.623(4)
680.71(9)
1
10.8697(1)
16.7084(3)
16.8867(4)
90
b (A)
˚
c (A)
a (8)
b (8)
g (8)
103.2599(8)
90
3
˚
V (A )
2985.12(9)
4
1.453
Z
rcalc (Mg mꢁ3
)
1.504
1.804
984
1.420
1.924
296
m (mmꢁ1
F(000)
)
1.937
1320
u Range (8)
Index ranges
3.04Á
135
215
19746
/
25.36
h 5
l 522
3.42Á
105
115
5010
/
25.31
h 5
l 511
3.03Á
145
215
27497
/
27.46
h 5
l 521
ꢁ
ꢁ/
/
/
/
13, ꢁ/135/k 5/14,
ꢁ
ꢁ/
/
/
/
9, ꢁ
/
105
/
k 5
/
10,
ꢁ
ꢁ/
/
/
/
14, ꢁ
/
215
/
k 521,
/
/
/
/
/
/
/
No. of reflections collected
No. of indep reflns/Rint
No. of refined parameters
8014/0.062
439
2401/0.046
117
6799/0.038
241
R1 (F) (I ꢁ
/
2s(I)); wR2 (F2) (all data) a
0.0372; 0.0662
0.921 and ꢁ0.757
0.0328; 0.0561
0.493 and ꢁ0.532
0.0365; 0.0784
1.019 and ꢁ0.892
ꢁ3
˚
Largest difference peak and hole (e A
)
/
/
/
a
R1ꢀ
/
ajjFojꢁ
/
jFcjj/ajFoj, wR2ꢀ
/
ꢀa w{(Fo)2ꢁ(Fc)2}/a w{(Fo)2}2.
/

338
P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á340
/
Table 2
axis of conformer 5A displays a distorted staggered
˚
˚
Selected bond lengths (A) and bond angles (8) for 1, 2 and 5
conformation with a tinÁ
/
tin distance of 2.8367(12) A
(2.8299(5) A for 6). The torsion angles are found
between 78.69(5) and 80.50(12)8 for Si(1)ÁSn(1)Á
Sn(2)ÁSi(2), C(11)ÁSn(1)ÁSn(2)ÁC(21) and Si(1)Á
Sn(1)ÁSn(2)ÁC(25), while the Si(1)ÁSn(1)ÁSn(2)Á
C(21), C(11)ÁSn(1)ÁSn(2)ÁC(25) and C(15)ÁSn(1)Á
Sn(2)ÁSi(2) torsion angles are found around 408.
˚
1
2
5
/
/
Bond lengths
/
/
/
/
/
Sn(1)Á
Sn(1)Á
Sn(2)Á
Sn(2)Á
Sn(1)Á
Si(1)Á
Si(1)Á
/
Si(1)
Si(3)
Si(1)
Si(2)
Sn(2)
2.572(2)
2.569(2)
2.577(2)
2.566(2)
2.610(2)
2.582(2)
/
/
/
/
/
/
/
/
/
/
/
/
/
/
2.547(2)
2.837(1)
2.065(2)
2.022(2)
/
/
Cl(1)
/
Cl(2)
3. Experimental
Bond angles
CÁ
CÁ
CÁ
CÁ
CÁ
Si(1)Á
Si(1)Á
Si(1)Á
C(15)Á
C(11)Á
CÁSi(1)Á
CÁSi(1)Á
CÁSi(1)Á
CÁSi(1)Á
CÁSi(1)Á
Sn(1)Á
Sn(1)Á
Sn(1)Á
C(21)Á
C(25)Á
/
Sn(1)Á
Sn(1)Á
Sn(1)Á
Sn(1)Á
Sn(1)Á
/
C
104.2(2)
107.0(1)
110.5(1)
114.0(1)
106.5(1)
111.8(2)
112.6(1)
115.3(1)
113.2(1)
112.2(1)
90.19(4)
109.63(15)
104.12(10)
107.08(9)
3.1. General methods
/
/
Si(1)
Si(1)
Si
/
/
/
/
All reactions were carried out under an atmosphere of
inert gas (N2 or Ar) using Schlenk techniques. All
solvents were dried by standard methods and freshly
distilled prior to use. All other chemicals used as starting
materials were obtained commercially. 1H- and 13C-
NMR spectra were recorded using Bruker DRX 400 or
DRX 500 spectrometer (internal reference Me4Si). 29Is-
and 119Sn-NMR spectra were recorded using a Bruker
DRX 400 spectrometer or a Bruker DPX 300 spectro-
meter (internal reference Me4Si or Me4Sn, respectively).
Mass spectra were obtained using a MAT 8200 mass
spectrometer. Elemental analyses were performed on a
LECO-CHNS-932 analyzer.
/
/Si
/
Sn(1)Á
Sn(1)Á
Sn(1)Á
Sn(1)Á
Sn(1)Á
/
Si(1)
Si(3)
Sn(2)
/
/
114.2(5)
/
/
111.92(3)
110.92(8)
112.82(10)
107.2(8)
/
/Sn(2)
/
/Sn(2)
/
/C
108.2(3)
109.9(2)
110.2(2)
107.8(2)
111.0(2)
105.9(2)
116.8(2)
114.1(2)
116.5(1)
113.7(2)
89.81(4)
/
/
Sn(1)
Sn(1)
Sn
114.23(13)
114.3(8)
/
/
/
/
/
/Sn
/
Si(1)Á
Si(1)Á
Si(1)Á
Si(1)Á
Si(1)Á
/
Sn(1)
Sn(2)
Cl(1)
/
/
109.77(6)
/
/
109.87(8)
106.97(8)
103.5(8)
/
/
Cl(1)
Chromatographic analysis of ethylene evolved in the
reaction between dibromoethane and bi-organometallic
compounds was performed on the chromatograph
‘Tsvet-104’ using 1 m glass column with activated
charcoal.
/
/
Cl(1)
often in organometallic chemistry mainly due to the
similar radii and electron density of the groups. In
addition, the tertiary carbon atoms of one t-butyl group
and the silicon atom of a chlorodimethylsilyl group are
sharing their positions exactly (EXYZ [18]), suggesting
that only small energetic differences were given between
the two conformers 5A (syn-clinal, skew or gauche
conformation) and 5B (anti-periplanar, staggered con-
formation) in the solid state. This result is in contrast to
the solid state structure of the related 1,2-dichlorotetra-
3.2. Kinetic investigations
The rate of magnesium oxidation by diphenyltin
dichloride was measured by controlling the electric
resistance of the magnesium wire (dꢀ
/0.5 mm, magne-
sium content ꢁ99.9%) put in the reaction mixture [14].
/
This method was modified for performing experiments
in an argon atmosphere. Before starting measurement
magnesium wire was activated in the reaction cell by
adding a 0.01 M solution of iodine in THF so that the
change of its resistance was about 10%. Afterwards the
iodine solution was removed from the reaction cell, the
cell was washed with pure THF twice, and the oxidizing
mixture was added.
tert-butyldistannane (t-Bu2Sn(Cl)ÁSn(Cl)t-Bu2, 6) for
which only a trans conformer similar to 5B was
observed [19] (Chart 3).
Therefore, the two pairs (Si(2), C(25?)) and (Si(2?),
C(25)) were refined with the same anisotropic displace-
/
ment parameters (EADP [22]). A view along the tinÁtin
/
Chart 3. Conformers of compound 5 in the solid state. 5A (syn-clinal, skew or gauche conformation) and 5B (anti-periplanar, staggered
conformation).

P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á
/
340
339
3.3. General procedures for 1 and 2: the
stannasilacycloalkanes 1 and 2 were prepared according
to the procedures described in Ref. [3]
the remaining solution in a refrigerator gave 3.95 g
1
(80%) of 5 as a colorless solid, mp (dec.): 206 8C. H-
NMR (500.13 MHz, C6D6):
3J(119SnÁ1 26 Hz], 1.49 [SnCMe3, J(119SnÁ1
H)ꢀ
65 Hz]. 13C{1H}-NMR (125.77 MHz, C6D6): d 8.7
[SiMe2, 2J(119SnÁ13
C)ꢀ37 Hz], 31.4 [SnC(CH3)3,
1J(119SnÁ13 253 Hz, 2J(119SnÁ13
C)ꢀ C)ꢀ16.3 Hz],
34.6 [SnC(CH3)3]. 29Si{1H}-NMR (59.63 MHz, C6D6):
d 31.3 [1J(119SnÁ29 310 Hz, 2J(119SnÁ29
Si)ꢀ Si)ꢀ51
Hz]. 119Sn{1H}-NMR (111.91 MHz, C6D6): d ꢁ
[1J(119SnÁ117
Sn)ꢀ73 Hz]. Anal. Calc.
d
0.77 [SiMe2,
3
/
/
/
H)ꢀ
/
3.3.1. 1,1,3,3,5,5-Hexamethyl-2,2,4,4,6,6-hexaphenyl-
1,3,5-trisila-2,4,6-tristannacyclohexane, [Á
/
SiMe2Á
/
/
/
SnPh2Á]3 (1)
/
/
/
/
/
Starting materials: 0.94 g (7.3 mmol) of Me2SiCl2, 2.5
g (7.3 mmol) of Ph2SnCl2, 70 ml of THF, 0.5 g (21
mmol) of Mg. The crude product was dissolved in 50 ml
of n-hexane, and filtered (G3, with 1 cm of Celite). After
/
/
/
/
/122.8
/
/
for
removal of ꢀ
refrigerator (5Á
m.p.ꢁ250 8C, were obtained. H-NMR (500,13 MHz,
C6D6): d 0.73 [SiMe2, 3J(119SnÁ1
H)ꢀ33 Hz], 7.58
[SnPh2]. 13C{1H}-NMR (125.77 MHz, C6D6): d ꢁ
0.7
[SiMe2, 2J(119SnÁ13
C)ꢀ16 Hz], 128.3 [SnPh2-para],
128.9 [SnPh2-meta, 3J(119SnÁ13
C)ꢀ40 Hz], 138.6
[SnPh2-ortho, 2J(119SnÁ13
C)ꢀ35 Hz], 139.4 [SnPh2-
ipso, 1J(119SnÁ13 not found, 3J(119SnÁ13
C)ꢀ C)ꢀ22
Hz]. 29Si{1H}-NMR (59.63 MHz, C6D6): d ꢁ
35.0
[1J(119SnÁ29 364 Hz]. 119Sn{1H}-NMR (111.91
Si)ꢀ
MHz, C6D6): d Si)ꢀ361 Hz,
212.2 [1J(119SnÁ29
2J(119SnÁ117
Sn)ꢀ827 Hz]. Anal. Calc. for C42H48Si3Sn3
/
40 ml of the solvent and storage in a
C20H48Si2Sn2Cl2 (653.23): C, 36.8; H, 7.4. Found: C,
37.1; H, 7.5%.
/
8 8C) 2.1 g (87%) of 1 as a colorless solid,
1
/
/
/
3.4. Crystallography of 1, 2 and 5
/
/
/
The crystals were mounted on the diffractometer in
sealed Lindemann capillaries. The data were collected at
173 K to a maximum u of 25.368 with 323 frames for 1,
of 25.318 with 249 frames for 2 and of 27.468 with 290
/
/
/
/
/
/
/
/
/
frames for 5 via v-rotation (D/vꢀ
frame on a Nonius Kappa CCD diffractometer using
graphite monochromated MoÁKa radiation (lꢀ
/
18) twice 10 s per
/
/
ꢁ
/
/
/
/
/
˚
/
/
0.71073 A). The structures were solved by direct
methods (SHELXS-97) [20] missing atoms, were located
in subsequent difference Fourier cycles and refined by
full-matrix least-squares of F2 (SHELXL-97) [21]. All non-
hydrogen atoms were refined using anisotropic displace-
ment parameters.
(995.01): C, 50.9; H, 4.9. Found: C, 50.4; H, 4.9%.
3.3.2. 1,1,3,3-Tetramethyl-2,2,4,4-tetra-tert-butyl-1,3-
disila-2,4-distannacyclobutane, [Á
/
SiMe2Á
/
SnÁ
/
t-Bu2Á]
/
2
(2)
Starting materials: 0.9 g (7 mmol) of Me2SiCl2, 2.0 g
(6.8 mmol) of t-Bu2SnCl2, 70 ml of THF, 0.9 g (37
mmol) of Mg. Further purification follows the proce-
The H atoms were placed in geometrically calculated
positions using a riding model with Uiso constrained at
1.2 for non-methyl and 1.5 for methyl groups times Ueq
of the carrier C atom.
Atomic scattering factors for neutral atoms and real
and imaginary dispersion terms were taken from the
International Tables for X-ray Crystallography [22].
The figures were created by SHELXTL [18].
dure described for compound 1 yielding 3.38 g (85%) of
1
2 as a colorless solid with a m.p.ꢁ
/
250 8C. H-NMR
3
(500.13 MHz, C6D6): d 0.83 [SiMe2, J(119SnÁ1
Hz], 1.44 [SnCMe3, (3J119SnÁ1 57 Hz]. 13C{1H}-
H)ꢀ
NMR (125.77 MHz, C6D6):
/
H)ꢀ26
/
/
/
d
0.97 [SiMe2,
[SnC(CH3)3,
[SnC(CH3)3
2J(119SnÁ13
1J(119SnÁ13
2J(119SnÁ13
MHz, C6D6):
119Sn{1H}-NMR (111.91 MHz, C6D6):
[1J(119SnÁ29
Si)ꢀ239 Hz]. Anal. Calc. for C20H48Si2Sn2
/
C)ꢀ
C)ꢀ
C)ꢀ
/
7
Hz],
found],
30.2
34.0
/
/not
3.5. Theoretical calculations
/
/
not observed]. 29Si{1H}-NMR (59.63
d
ꢁ
/
9.1 [1J(119SnÁ29
/
Si)ꢀ/239 Hz].
The reaction enthalpy of a given reaction is deter-
mined by the sum of the standard heat of formation of
the components. These standard heat of formation are
not available directly from ab-initio calculations, there-
fore, the group equivalent method from Wiberg and
d
ꢁ
/
40.5
/
/
(582.18): C, 41.3; H, 8.3. Found: C, 40.8; H, 8.2%.
3.3.3. 1,2-bis(Dimethylchlorosilyl)-tetra-tert-
coworkers was used [23Á25]. This method describe a
/
butyldistannane, ClÁ
/
SiMe2Á
/
t-Bu2SnÁ
/
t-Bu2SnÁ
/
Me2SiÁ
/
link between experimental determined enthalpies of
formation and ab-initio energies. One of the major
problems is the quantification of the ring strain in cyclic
systems. However, experimental determined heats of
formation are only little known for cyclic silanes or
stannanes and so a model for the description of the ring
strain must be used. For the description of the ring
strain are a number of models known. All of them are
based on the difference of energies between strained and
unstrained rings [26]. With this assumption and the
Cl (5)
Me2SiCl2 (1.0 g, 7.7 mmol), 2.3 g (7.7 mmol) t-
Bu2SnCl2, and 0.5 g (37 mmol) Mg are stirred in 80 ml
THF at room temperature (r.t.). After 5Á10 min the
/
mixture turned from colorless to black which indicated
the start of the reaction. After 1 h the solvent was
removed in vacuo and 50 ml n-hexane was added. The
solution was filtered (G3, with 1 cm of Celite) and
ꢀ90% of the solvent was removed in vacuo. Storage of
/

340
P. Bleckmann et al. / Journal of Organometallic Chemistry 686 (2003) 332Á340
/
[5] U. Englich, K. Ruhlandt-Senge, F. Uhlig, J. Organomet. Chem.
613 (2000) 139.
known Wiberg group equivalents [27] the energies of
strained cycles are calculable in comparison to those of
unstrained cyclic systems to yield a correction for the
strained group equivalents [27,28].
[6] M. Westerhausen, Angew. Chem. Int. Ed. Engl. 33 (1994) 14.
[7] Organometallic Compounds and Radicals, Nauka, Moscow,
1985, p. 204 (in Russian).
[8] U. Blaukat, W.P. Neumann, J. Organomet. Chem. 63 (1973)
27.
The strictness of the resulting energies of formation
are within the scope of experimental limits [26,27,29].
[9] W.P. Neumann, J. Pedain, R. Sommer, Justus Liebigs Ann.
Chem. 694 (1966) 9.
The ab-initio calculations are based on HartreeÁFock
/
with Stuttgarter basis set [30] by using the program
package GAUSSIAN-98 [31].
[10] B. Josseaume, N. Noiret, M. Pereyre, A. Saux, Organometallics
13 (1994) 1034.
[11] C. Tamborski, E.J. Soloski, J. Am. Chem. Soc. 83 (1961) 3734.
[12] H.J.M. Creemers, J.G. Noltes, G.J. Van der Kerk, J. Organomet.
Chem. 14 (1968) 217.
4. Supplementary material
[13] J.-C. Lahournere, J. Valade, J. Organomet. Chem. 22 (1970)
C3.
[14] J.F. Garst, F. Ungvary, in: H.G. Richey, Jr (Ed.), Grignard
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC nos. 206079, 206078 and 206080
for compounds 1, 2 and 5, respectivley. Copies of this
information may be obtained free of charge from The
Director, CCDC, 12 Union Road, Cambridge CB2 1EZ,
Reagents*New Developments, Wiley, Chichester, 2000, p. 185.
/
[15] (a) S.A. Zhukov, I.P. Lavrentyev, T.A. Nifontova, React. Kinet.
Catal. Lett. 4 (1974) 1105;
(b) A.V. Piskounov, S.V. Maslennikov, I.V. Spirina, V.P.
Maslennikov, Appl. Organometal. Chem. 14 (2000) 570 (and
references therein).
[16] G. Bremer, K.-P. Wendlandt, Introduction into Heterogeneous
Catalysis (Russian ed.), Moscow, Mir, 1981, p. 48 (Russian ed.).
[17] F. Uhlig, G. Reeske, U. Hermann, M. Schurmann, Z. Anorg.
¨
UK (Fax: ꢂ44-1223-336033; e-mail: deposit@ccdc.
cam.ac.uk or www: http://www.ccdc.cam.ac.uk).
/
Allg. Chem. 627 (2001) 453.
[18] G.M. Sheldrick, SHELXTL, Release 5.1 Software Reference
Manual, Bruker AXS Inc., Madison, WI, 1997.
Acknowledgements
[19] F. Uhlig, I. Prass, U. Hermann, T. Schollmeier, U. Englich, K.
Ruhlandt-Senge, J. Organomet. Chem. 646 (2002) 271.
[20] G.M. Sheldrick, Acta Crystallogr. Sect. A 46 (1990) 467.
The authors thank funds from the Deutsche For-
schungsgemeinschaft (DFG, Germany) and the Fonds
der chemischen Industrie (Germany). We also acknowl-
edge Prof. Dr. K. Jurkschat (Dortmund University) for
his interest in this work.
[21] G.M. Sheldrick, SHELXL-97, University of Gottingen, Germany,
¨
1997.
[22] International Tables for Crystallography, vol. C, Kluwer Aca-
demic Publishers, Dordrecht, 1992.
[23] K.B.J. Wiberg, Comp. Chem. 5 (1984) 197.
[24] M.R. Ibrahim, P. von Rague´ Schleyer, J. Comp. Chem. 6 (1985)
157.
References
[25] K.B.J. Wiberg, Org. Chem. 50 (1985) 5285.
[26] K.B.J. Wiberg, Angew. Chem. 98 (1986) 312.
[27] T. Bruggemann, Diploma thesis, Dortmund University, 2000 and
[1] F. Uhlig, R. Hummeltenberg, K. Jurkschat, Phosphorus Sulfur
Silicon Relat. Elem. 123 (1997) 255.
[2] F. Uhlig, C. Kayser, R. Klassen, U. Hermann, L. Brecker, M.
¨
references therein.
[28] T. Bruggemann, P. Bleckmann, Unpublished results, 2002.
¨
Schurmann, K. Ruhlandt-Senge, U. Englich, Z. Naturforsch. Teil
¨
b 54 (1999) 278.
[29] M. Braunschweig, PhD thesis, Dortmund University, 2000.
[30] H. Fuentealba, H. Preuss, H. Stoll, L. v. Szentpaly, Chem. Phys.
Lett. 89 (1989) 418.
[3] C. Kayser, R. Klassen, M. Schurmann, F. Uhlig, J. Organomet.
¨
Chem. 556 (1998) 165.
[31] J.A. Pople, M.J. Frisch, G.W. Trucks, GAUSSIAN-98 Program
Package, Gaussian Inc., Pittsburgh, PA, 1998.
[4] M. Schurmann, F. Uhlig, Organometallics 21 (2002) 986.
¨