Trovacene chemistry. 16 [1] 1,1 a-Di([5]trovacenyl)-1,1,1a,1a-tetraphenyl- disiloxane: Electro- and magneto-communication across a Si-O-Si spacer

Article abstract of DOI:10.1002/zaac.200700275

By means of lithiation and subsequent derivatization [5]trovacenyl- diphenylsilanol (2^.), di([5]trovacenyl)tetraphenyl-disiloxane (3 ^..), di([5]trovacenyl)tetramethyl-disiloxane (4^..) and di([5]trovacenyldimethylsilyl)methane (6^..) have been prepared and characterized by X-ray diffraction (2^., 3^.., 6 ^..). Although the bond angles of the disiloxane bridges differ considerably, magnetic exchange interaction in these organometallic diradicals is practically unaffected as shown by fluid solution EPR spectroscopy. By way of contrast, electrocommunication clearly responds to change in the structure of the SiOSi spacer in that the redox splittings derived from cyclic voltammetry display the gradation δE_1/2 (0/-/2-) 4^.. < 6 ^.. < 3^... It is proposed that in this order electronic polarizability in the intervening spacer increases thereby facilitating the transmission of electrostatic effects.

Full text of DOI:10.1002/zaac.200700275

DOI: 10.1002/zaac.200700275
Trovacene Chemistry. 16 [1]
1,1a-Di([5]trovacenyl)-1,1,1a,1a-tetraphenyl-disiloxane:
Electro- and Magneto-Communication Across a Si-O-Si Spacer
Christoph Elschenbroicha,*, Feng Lua, Olaf Burghausa, Klaus Harmsa, and Mathias Nowotnyb
a Marburg, Fachbereich Chemie der Philipps-Universität
b Darmstadt, Institut für Anorganische Chemie der Technischen Hochschule
Received May 24th, 2007.
Dedicated to Professor Dieter Fenske on the Occasion of his 65th Birthday
Abstract. By means of lithiation and subsequent derivatization
[5]trovacenyl-diphenylsilanol (2.), di([5]trovacenyl)tetraphenyl-di-
siloxane (3..), di([5]trovacenyl)tetramethyl-disiloxane (4..) and di([5]-
trovacenyldimethylsilyl)methane (6..) have been prepared and
characterized by X-ray diffraction (2., 3.., 6..). Although the bond
angles of the disiloxane bridges differ considerably, magnetic ex-
change interaction in these organometallic diradicals is practically
unaffected as shown by fluid solution EPR spectroscopy. By way
of contrast, electrocommunication clearly responds to change in
the structure of the SiOSi spacer in that the redox splittings derived
from cyclic voltammetry display the gradation δE1/2 (0/Ϫ/2Ϫ) 4..
< 6.. < 3... It is proposed that in this order electronic polarizability
in the intervening spacer increases thereby facilitating the trans-
mission of electrostatic effects.
Keywords: Trovacene; Silanol; Disiloxane; Spin exchange coupling;
Cyclic voltammetry
1 Introduction
(51V). The work described here is part of our systematic
investigation of electron spin-spin coupling and redox split-
ting in oligotrovacenyls as a function of the nature of the
spacers [4].
Inorganic and organometallic polymers have received atten-
tion with regard to the possibility that they exhibit interest-
ing materials properties such as rigidity, electrical conduc-
tivity, non-linear optical phenomena, or molecular ferro-
magnetism [2]. Whereas backbones composed of main
group nonmetals (polysilanes, polysiloxanes, polyphosphaz-
enes, polyborazylenes) have a long tradition metal-contain-
ing polymers, are of more recent vintage [2e, k]. The latter
comprise two classes, namely polymers that incorporate or-
ganometallic units in their backbone, and those in which
these units form substituents at the polymer chain. In the
following we disclose the synthesis, structural elucidation,
EPR spectra, magnetic susceptometry and cyclic voltamme-
try of a disiloxane which bears two paramagnetic [5]trova-
cenyl groups [3]. This molecule may be regarded as the
smallest representative in the class of poly([5]trovacenyl)si-
loxanes and should lend itself to a study of the extent of
intramolecular intermetallic communication mediated by a
Si-O-Si bridge. Polysiloxanes decorated with paramagnetic
sandwich units could lead to super paramagnetic nanos-
tructures [2b]; they also could serve in studies of motional
properties since the latter modulate the 51V hyperfine tensor
with attendant effects on the linewidths as a function of mI
2 Results and Discussion
The most common access to disiloxanes proceeds via
H2O
the sequence R3SiCl
Ǟ R3SiOH 2 ϫǞ R3SiOSiR3.
ϪH2O
Herein, the ease of dehydration is controlled by the nature
of R. Typically, the condensation of triphenylsilanol calls
for concentrated acid or for strongly alkaline media and
elevated temperature [5, 6], conditions which are detrimen-
tal to the trovacene unit. We nevertheless prepared [5]trova-
cenyldiphenylsilanol (2.) (Scheme 1), a potential precursor
to di([5]trovacenyl)tetraphenyl- disiloxane, because silanols
present remarkable structural features imposed by strong
hydrogen bonding [6]. The molecular structure of the sil-
anol 2. crystallized from ethanol is presented in Figure 1,
bond lengths and bond angles are collected in Table 1. Con-
trary to triphenylsilanol, which in the crystal forms hydro-
gen bonded tetramers [5], [5]trovacenyldiphenylsilanol gen-
erates hydrogen bonded dimers, which incorporate an ad-
ditional molecule of ethanol.
The larger size of the trovacenyl compared to the phenyl
substitutent at silicon is an obvious explanation for the
lower degree of aggregation in 2. compared to triphenylsil-
* Prof. Dr. C. Elschenbroich
Fachbereich Chemie der Universität
D-35032 Marburg
˚
anol. Considering the O-H···O distances (2.74, 2.74 A) and
E-mail: eb@chemie.uni-marburg.de
angles (167°, 170°), the hydrogen bonds in 2. are moderately
2386
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 633, 2386Ϫ2394

Trovacene Chemistry. 16
˚
Table 1 Selected bond lengths /A and angles /° for [5]trovacenyl-
diphenylsilanol 2. as found for (2.)2 ·C2H5OH.
Si1ϪC1
1.861(3)
1.869(4)
1.877(3)
1.869(3)
2.275(3)
2.260(3)
2.268(3)
2.255(4)
2.247(4)
V1ϪC6
V1ϪC7
V1ϪC8
V1ϪC9
V1ϪC10
V1ϪC11
V1ϪC12
V1ϪC7 ring (mean)
V1ϪC7 (centroid)
C6ϪC7
C6ϪC12
C7ϪC8
C8ϪC9
C9ϪC10
C10ϪC11
C11ϪC12
CϪC (mean for C7)
O1ϪH1 ···O1Ј
2.171(4)
2.172(4)
2.180(5)
2.188(5)
2.185(5)
2.178(4)
2.172(4)
2.178(4)
1.468(1)
1.382(7)
1.414(8)
1.410(7)
1.394(7)
1.372(8)
1.394(8)
1.406(7)
1.396(8)
2.742(3)
2.740(4)
Si1ϪC13
Si1ϪC19
Si1ϪC (mean)
V1ϪC1
V1ϪC2
V1ϪC3
V1ϪC4
V1ϪC5
V1ϪC5 ring (mean) 2.261(3)
V1ϪC5 (centroid) 1.913(2)
Si1ϪO1
O1ϪH1
C1ϪC2
C1ϪC5
C2ϪC3
C3ϪC4
C4ϪC5
1.626(2)
0.78(5)
1.433(5)
1.426(5)
1.401(5)
1.406(6)
1.422(5)
···
CϪC (mean for C5) 1.417(5)
O1ЈϪH1Ј O100
O1-Si1-C1
109.85(15)
109.21(15)
109.31(15)
110.50(14)
110.63(15)
107.29(15)
120(3)
128.6(3)
126.3(3)
121.5(3)
105.0(3)
116.6(3)
116.8(3)
122.3(3)
167
O1-Si1-C1-C5
45.8(3)
165.6(3)
Ϫ76.5(3)
Ϫ137.5(3)
Ϫ177.4(3)
0.0(4)
Ϫ172.7(3)
67.1(3)
Ϫ52.9(3)
179.7(3)
Ϫ1.5(6)
O1-Si1-C13
C1-Si1-C13
O1-Si1-C19
C1-Si1-C19
C13-Si1-C19
Si1-O1-H1
C5-C1-Si1
C2-C1-Si1
C14-C13-Si1
C5-C1-C2
C13-Si1-C1-C5
C19-Si1-C1-C5
O1-Si1-C1-C2
Si1-C1-C2-C3
C5-C1-C2-C3
O1-Si1-C13-C14
C1-Si1-C13-C14
C19-Si1-C13-C14
Si1-C13-C14-C15
C13-C14-C15-C16
O1-Si1-C19-C20
C1-Si1-C19-C20
Torsion C7 ring (mean)
C14-C13-C18
C20-C19-C24
C20-C19-Si1
Ϫ149.3(3)
Ϫ27.4(3)
0.0(0)
Figure 1 (a) Molecular structure of [5]trovacenyl-diphenylsilanol
(2.) in a crystal grown from ethanol (50 % probability ellipsoids).
···
O1ϪH1 O1Ј
Torsion C5 Rring (mean)
0.0(0)
Two units of 2.connected by a HSiO-H···OSi hydrogen bond and
···
O1ЈϪH1Ј O100 170
H
interacting with a C2H5OH molecule are shown. (b) Stereoview of
the unit cell illustration the packing of the dimers (2.)2 ·C2H5OH.
transition metal derivative of triphenylsilanol compound 2.
resembles Ph2[η6-PhCr(CO)3]Si(OH); for the latter the
siloxy hydrogen atom could not be precisely located and
intermolecular hydrogen HSiO-H···OSi bonding was not
strong [7]. Interestingly, despite differences in environment
the hydrogen bonds O(1)-H(1)···O(1Ј) and O(1Ј)-
H(1Ј)···O(100) are of almost equal strength. A totally differ-
ent packing of 2. is encountered if crystals are grown from
benzene/petroleum ether (Figure 2, Table 2). No solvent
molecules are incorporated and significant intermolecular
hydrogen bonding is absent in this case.
H
address [9].
Since dehydration of 2. failed, we prepared the target
compound 3.. by the path given in Scheme 1. The side prod-
uct [5]trovacenyldiphenylsiloxy-diphenylsilanol 5. probably
stems from single C-Si coupling and hydrolysis of the re-
maining Si-Cl bond during workup. X-ray crystallography
performed on crystals of 5. yielded the structure which is
depicted in Figure 3; it merely served unequivocal product
identification and will not be discussed in detail here.
The structure of the title compound 1,1a-di([5]trova-
cenyl)-1,1,1a,1a-tetraphenyldisiloxnae (3..) is shown in Fig-
ure 4, bond lengths and angles are listed in Table 3. Obvi-
ously the structure of 3.. should be compared to that of
hexaphenyldisiloxane (7). The latter has been reported to
be centrosymmetric, the SiOSi fragment being strictly linear
At best very weak interactions O-H···Cπ, C-H···Cπ can
be spotted between silanol- and cycloheptatrienyl hydrogen
atoms and the π-perimeter of a neighboring η7-C7H7 li-
gand; they are marked in Figure 2. Thus, as taken from
Table 2, the distances d(H···C) are only slightly smaller than
˚
the sum of van der Waals radii for C and H (ഠ 2.90 A [7]).
The fact that the more acidic silanol OH function gives rise
to a smaller d(H···C) value than the η7-C7H7 donor, may
be regarded as evidence for the existence of hydrogen bond-
ing to Cπ acceptors [8].
The absence of O-H···O hydrogen bonding for crystalline
2. contrasts with triphenylsilanol which forms a hydrogen
bonded cyclic tetramer (Ph3SiOH)4 and additionally incor-
porates one molecule of ethanol yielding a pentagon of hy-
drogen bonded oxygen atoms (Ph3SiOH)4 ·EtOH [5]. As a
˚
and featuring the bond length Si-O ϭ 1.616 (1) A [10]. The
SiOSi backbone in 3.. differs in that it is slightly bent
[168.4(2)°], correspondingly, the Si-O bonds are a little
˚
longer [1.6212(7) A] than those in 7. If the SiOSi segment
in 3.. is approximated to be linear, the structure may be
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
2387

C. Elschenbroich, F. Lu, O. Burghaus, K. Harms, M. Nowotny
Scheme 1
designated syn clinal. With regard to the Si-Cipso bonds, the
lation furnished the value ΗJEPR(3..)Η ϭ 0.070 cmϪ1 for the
exchange parameter. This corresponds to ΗJΗ ϭ 75 mT and
implies, that ΗJEPR(3..)Η ഠ 10 a(51V, 1.). If one strives at put-
ting the exchange coupling constant JEPR(3..) in perspective,
comparing it with values obtained for hydrocarbon spacers,
Z-2,3-di([5]trovacenyl)-2-butene (9..) is the species of choice
in that ΗJEPR(9..)Η ϭ 0.052 cmϪ1 [14].
Accordingly the three-atom disiloxane spacer matches
the two atom Z-alkene unit in the ability to mediate elec-
tron-spin exchange-coupling. As a caveat it must be stated
that it is not only the chemical nature of the spacer but also
steric particularities which govern the extent of exchange
coupling. This is illustrated by the fact that for the isomer
E-2,3-di([5]trovacenyl)-2-butene (10..) exchange coupling is
much more extensive ΗJ(10..)Η ഠ 20 ΗJ(9..)Η. This has been at-
tributed to the greater ease of 10.. compared to 9.. to achieve
coplanarity of the cyclopentadienyl- and alkene π-electron
systems. The dramatic difference in exchange-coupling
efficiency for the isomers 9.. and 10.. strongly suggests that
π-delocalization plays an important role, as the inter-
vanadium distances are very similar and the number of in-
tervening bridge atoms are identical for 9.. and 10...
The extensive SiOSi bond angle widening to approach
linearity and the short SiO bonds in 3.. suggest the presence
of two neighboring SiϭO double bonds. Therefore com-
pound 3.. bears strong resemblance to the [2]cumulene 1,3-
diferrocenylallene (R)FcCϭCϭCFc(R) in that the spacer
features two orthogonal π- bonds [15]. As far as π-bonding
is concerned, odd cumulenes with C3, C5, C7 backbones are
nonconjugated and electronically decoupled. It therefore
does not come as a surprise that exchange coupling in 3..
(ΗJEPRΗ ϭ 0.070 cmϪ1) is much weaker than that in 1,4-di[5]-
trovacenylbutadiyne (11..) (ΗJEPRΗ ϭ 0.56 cmϪ1) [16] despite
[5]trovacenyl groups assume conformations which maximize
···
˚
the V1 V1a distance [7.85 A] thereby minimizing inter-tro-
vacene repulsion. Intermolecular forces must also be
reckoned with, however, as has been demonstrated for the
closely related 1,1a-diferrocenyl-1,1,1a,1a-tetramethyldis-
iloxane (8) which exhibits a nearly eclipsed conformation
when viewed along the SiOSi axis [angle 159.9(2)°] whereas
a staggered form is adopted in a cocrystal 8·ferrocene
[SiOSi 141.5(1)°] [11]. This study once again illustrates the
flexibility of the disiloxane linkage and suggests, that the
structure of 3.. in solution may differ considerably from that
in the solid state.
The methyl analog of 3.., namely 1,1a-di([5]trovacenyl)-
1,1,1a,1a-tetramethyldisiloxane (4..) is expected to display
structural differences of the Si-O-Si segment which could
have an impact on intramolecular communication; 4.. was
therefore also synthesized and subjected to an X-ray dif-
fraction study (Figure 5, Table 4). The most important fea-
ture of 4.. is the more strongly bent disiloxane bridge (θ
SiOSi ϭ 146.6°).
Intramolecular magnetocommunication in 3.. manifests it-
self most clearly in the EPR hyperfine pattern observed in
fluid solution (Figure 6); 15 components separated by a(51V,
3..)/2, which strongly deviate in intensity from the
1:2:3···8:7···1 distribution attest to the region of intermedi-
ate exchange on the a(51V) time scale [14]. Spectral simu-
2388
www.zaac.wiley-vch.de
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394

Trovacene Chemistry. 16
˚
Table 2 Selected bond lengths /A and angles /° for [5]trovacenyldi-
phenylsilanol (2.) as found for crystals devoid of included solvent
molecules.
C1ϪSi2
C13ϪSi2
C19ϪSi2
Si1ϪC(mean)
C1ϪV1
C2ϪV1
C3ϪV1
C4ϪV1
C5ϪV1
C5ϪV1 ring (mean) 2.252(2)
O1ϪSi2
O1ϪH1
C1ϪC2
C1ϪC5
C2ϪC3
C3ϪC4
C4ϪC5
d(H···C)
1.856(2)
1.874(2)
1.865(2)
1.865(2)
2.270(2)
2.248(2)
2.249(2)
2.246(2)
2.247(2)
C6ϪV1
C7ϪV1
C8ϪV1
C9ϪV1
C10ϪV1
C11ϪV1
C12ϪV1
V1ϪC7 ring (mean)
C6ϪC12
C6ϪC7
C7ϪC8
C8ϪC9
C9ϪC10
C10ϪC11
C11ϪC12
CϪC (mean for C7)
CϪC (mean for C5)
2.169(2)
2.171(2)
2.179(2)
2.178(2)
2.177(2)
2.174(2)
2.177(2)
2.175(2)
1.395(3)
1.388(3)
1.373(3)
1.389(3)
1.403(4)
1.409(4)
1.412(4)
0.0(0)
1.6360(13)
0.74(3)
1.426(2)
1.430(3)
1.403(3)
1.404(3)
1.413(3)
2.75(3)
0.0(0)
O1-Si2-C1
109.32(8)
106.80(7)
109.05(7)
108.99(8)
112.63(7)
109.88(7)
118(3)
126.00(13)
128.28(13)
123.02(12)
109.5(2)
C20-C19-Si2
C5-C1-Si2-O1
C5-C1-Si2-C13
C5-C1-Si2-C19
C2-C1-Si2-O1
C14-C13-Si2-C1
C5-C1-C2-C3
C14-C13-Si2-O1
C24-C19-Si2-C13
Si2-C13-C14-C15
C13-C14-C15-C16
C20-C19-Si2-O1
C20-C19-Si2-C1
Torsion C7 ring (mean)
Torsion C5 ring (mean)
119.28(14)
14.58(14)
Ϫ101.84(14)
135.90(14)
Ϫ170.99(14)
Ϫ135.57(13)
0.1(2)
106.41(14)
Ϫ64.4(2)
Ϫ179.30(13)
1.1(3)
O1-Si2-C13
C1-Si2-C13
O1-Si2-C19
C1-Si2-C19
C19-Si2-C13
Si2-O1-H1
C5-C1-Si2
C2-C1-Si2
C14-C13-Si1
C3-C1-C2
C14-C13-C18
C20-C19-C24
Si2-C1-C5-C4
117.1(2)
117.5(2)
175.35(12)
2.9(2)
Ϫ118.60(14)
0.0(0)
0.0(0)
would lead to a rise in exchange coupling, we prepared
compound 4.. (Scheme 1) which features the angle θ
(SiOSi) ϭ 146.6° (Figure 5, Table 4). Surprisingly the fluid
solution EPR spectrum of 4.. (Figure 6) is superimposable
on that of 3.. (ΗJ(4..)Η ϭ 0.076 cmϪ1 ഠ 81 mT, i.e. the pro-
nounced difference in SiOSi angles does not lead to a mag-
netostructural effect. If one would be pressed to come up
with an explanation one could argue that upon decreasing
the SiOSi angles two contributions cancel: (1) proceeding
from the 3 atom hetero-cumulene electronic structure (180°)
to a participation of an allyl structure (146.6°) increases π
conjugation and exchange coupling mediated by a π path;
(2) the decrease in SiOSi angle will decrease the s character
of the σ bonds thereby weakening magnetic communication
transmitted by a σ path. A different type of reasoning is
based on the fact that the SiOSi angle bending potential in
the range 130° < θ < 230° is very flat (inversion barrier
1.34 kJ molϪ1 [17]). This will serve to eliminate the struc-
tural differences observed for the solid state in that ther-
mally activated inversion of the SiOSi bridge can occur, ir-
respective of the substitutent at the Si atoms. Steric effects
will therefore not become manifest. Finally, the notion of
replacing the lone pairs at the central oxygen atom of the
disiloxane bridge by σ-bonding pairs led us to the synthesis
of di[[5]trovacenyldimethylsilyl]methane (6..) (Scheme 1)
Figure 2 (a) Molecular structure of [5]trovacenyldiphenylsilanol
(2.) in a crystal grown from petroleum ether/benzene (50 % proba-
bility ellipsoids). Weak intermolecular hydrogen bonds are indi-
cated by dotted lines. (b) Stereoview of the unit cell.
of the fact that in 11.. a C4 spacer is encountered whereas
3.. contains a C3 spacer. Conjugation of two neighboring
triple bonds is of course uninhibited and rotationally in-
variant. Supposedly the as yet unknown [3]cumulene spac-
ered dinuclear complex (C7H7)V(C5H4-CHϭCϭCϭCH-
H4C5)V(C7H7) will surpass 3.. in its propensity to mediate
exchange interaction because it features 1,3-butadiene con-
jugation.
Disiloxanes are known to differ in their Si-O-Si angles θ
dependent on the substituents at the Si atoms. Whereas for
Ph3SiOSiPh3 θ ϭ 180° applies [10], in siloxanes bearing less
bulky methyl groups the bridge deviates from linearity (ex-
ample: FcMe2SiOSiMe2Fc, θ (SiOSi) ϭ 159.9° [11]). In or-
der to explore whether lifting the orthogonality of the two
SiO double bond contributions in a nonlinear disiloxane
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
2389

C. Elschenbroich, F. Lu, O. Burghaus, K. Harms, M. Nowotny
˚
Table 3 Selected bond lengths /A and angles /° for 1,1a-Di([5]tro-
vacenyl)-1,1,1a,1a-tetraphenyl-disiloxane (3..).
C1ϪSi1
1.855(3)
1.6210(7)
1.6210(7)
1.872(2)
1.861(3)
2.281(2)
2.256(2)
2.257(3)
2.252(3)
2.259(2)
2.256(3)
2.185(3)
2.175(3)
2.182(3)
2.178(3)
2.177(3)
7.8532(7)
C1ϪC2
C1ϪC5
C2ϪC3
C3ϪC4
C4ϪC5
CϪC (mean for C5)
C6ϪC7
C6ϪC12
C7ϪC8
C8ϪC9
C9ϪC10
C10ϪC11
C11ϪC12
V1ϪC11
V1ϪC12
V1ϪC7 ring (mean)
1.423(4)
1.433(4)
1.399(4)
1.411(4)
1.403(4)
1.413(4)
1.413(6)
1.402(5)
1.385(6)
1.409(5)
1.389(5)
1.396(5)
1.398(4)
2.183(3)
2.183(4)
2.180(4)
O1ϪSi1a
O1ϪSi1
C13ϪSi1
C19ϪSi1
V1ϪC1
V1ϪC2
V1ϪC3
V1ϪC4
V1ϪC5
V1ϪC5 ring (mean)
V1ϪC6
V1ϪC7
V1ϪC8
V1ϪC9
V1ϪC10
···
V1 V2
C2-C1-Si1
C5-C1-Si1
125.6(2)
128.5(2)
C2-C1-C5
Si1a-O1-Si1
105.5(2)
168.5(2)
O1-Si1-C19
O1-Si1-C1
C1-Si1-C13
C5-C1-Si1-C19
Si1a-O1-Si1-C19
Si1a-O1-Si1-C13
107.37(12)
108.11(11)
109.99(11)
177.1(2)
Ϫ154.90(9)
88.82(10)
C1-Si1-C19
O1-Si1-C13
C19-Si1-C13
C2-C1-Si1-C13
Si1-C19-C24-C23
Si1a-O1-Si1-C1
114.88(11)
108.48(8)
107.83(12)
Ϫ113.3(2)
Ϫ178.9(3)
Ϫ30.41(8)
Figure 3 Molecular structure of [5]trovacenyldiphenylsiloxy-di-
phenylsilanol (5.) in the crystal (50 % probability ellipsoids). Para-
˚
meters for the disiloxane bridge: Si1ϪO1 1.635(2) A; Si2ϪO1
˚
˚
1.622(2) A; Si2ϪO2 1.618(2) A; Si1-O1-Si2 155.03(11)°.
Figure 5 Molecular structure of di[5]trovacenyltetramethyl-disilo-
xane (4..) in a crystal grown from petroleum ether / benzene (50 %
probability ellipsoids).
and its EPR study (Figure 6). Although no X-ray diffrac-
tion study was carried out for 6.. , it can be assumed with
certainty that the bridging angle Si-C-Si will be close to
tetrahedral (109.5°). The EPR spectrum of 6..in fluid solu-
tion and its simulation supplied the exchange parameter
ΗJ(6.. )Η ϭ 0.076 cmϪ1 (81 mT) which differs insignificantly
from those found for 3.. and 4... Apparently, the siloxane
Si-O-Si and carbosilane Si-CH2-Si spacers are equally effec-
tive in promoting exchange coupling in that O (lone pair)-
Si conjugation in 3.. and 4.. is replaced by CH-Si hypercon-
jugation in 6...
Figure
4
Molecular structure of 1,1a-Di([5]trovacenyl)-
1,1,1a,1aϪtetraphenyl-disiloxane (3..) in a crystal grown from
petroleum ether/benzene (50 % probability ellipsoids).
2390
www.zaac.wiley-vch.de
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394

Trovacene Chemistry. 16
˚
Table 4 Selected bond lengths /A and angles /° di[5]trovacenyl-
tetramethyl-disiloxane 4...
C1ϪC2
C1ϪC5
C2ϪC3
C3ϪC4
C4ϪC5
1.424(4)
1.424(4)
1.401(4)
1.389(5)
1.411(4)
C1ϪSi1
O1ϪSi1
O1ϪSi2
C13ϪSi2
V1ϪC(1)
V1ϪC2
1.855(3)
1.637(2)
1.631(2)
1.857(3)
2.256(3)
CϪC (mean for C5) 1.409(4)
2.241(3)
C6ϪC7
C6ϪC12
C7ϪC8
1.394(5)
1.395(5)
1.407(5)
1.387(6)
1.399(6)
1.396(6)
1.388(5)
2.177(2)
2.178(3)
2.170(3)
2.167(3)
2.175(3)
2.182(3)
2.180(3)
2.177(3)
8.8206(9)
92.1(2)
V1ϪC3
V1ϪC4
V1ϪC5
2.256(2)
2.267(3)
2.256(3)
2.255(3)
108.33(10)
126.5(2)
146.61(13)
76.1(3)
105.6(2)
127.7(2)
108.44(14)
107.11(12)
110.32(14)
Ϫ164.1(2)
Ϫ43.6(3)
Ϫ81.1(2)
C8ϪC9
V1ϪC5 ring (mean)
O1-Si1-C1
C5-C1-Si1
Si2-O1-Si1
Si2-O1-Si1-C26
C2-C1-C5
C2-C1-Si1
O1-Si1-C26
O1-Si1-C25
C26-Si1-C1
Si2-O1-Si1-C25
Si2-O1-Si1-C1
C5-C1-Si1-O1
C9ϪC10
C10ϪC11
C11ϪC12
V1ϪC6
V1ϪC7
V1ϪC8
V1ϪC9
V1ϪC10
V1ϪC11
V1ϪC12
V1-C7 ring (mean)
V1···V2
C2-C1-Si1-O1
Exchange coupling in oligotrovacenyls can be studied by
two methods which are complementary with regard to the
strength of this magnetic interaction. Owing to the magni-
tude of hyperfine splitting in parent trovacene [Ηa(51V, 1.)Η ϭ
6.98 mT ϭ 0.0065 cmϪ1] EPR analysis can be applied for
0.001 Շ ΗJΗ Շ 1 cmϪ1 whereas magnetic susceptometry re-
quires ΗJΗ տ 1 cmϪ1. Therefore, the very small exchange
coupling constants determined for 3.., 4.., and 6.. by fluid
solution EPR preclude an application of magnetic suscepto-
metry. It should be mentioned that in cases of exceedingly
weak intramolecular exchange coupling reliably quantified
by EPR, the interpretation of χ-T traces is complicated by
the problem of disentangling intra- and intermolecular con-
tributions. Quite frequently, in the solid state intermolecular
V···V distances are shorter than the intramolecular ones
and the χ-T curve then reflects a magnetic exchange inter-
action whose origin is not strictly intramolecular. This
problem has been dealt with previously in the case of
1,4-di([5]trovacenyl)butadiyne (C7H7)V(C5H4(-CϵC-CϵC-
(C5H4)V(C7H7) [16]. It may even be argued that in the solid
state, the differentiation intra/intermolecular becomes
blurred and we therefore refrain from presenting here the
χ-T curves for 3.., 4.., and 6.. and their multiparameter fits.
Electrocommunication in oligonuclear complexes may be
gauged from redox splittings, i.e, the differences in the
potentials of successive electron transfer steps. The cyclic
voltammograms (CV) for 3.., 4.. and 6.. are presented in
Figure 7, relevant data are collected in Table 5. CV traces
for oligonuclear trovacenes usually show three clearly sepa-
rated regions: vanadium centered reduction (Ϫ3.0 Շ E Շ
Ϫ2.4 V), vanadium-centered oxidation (0 Շ E Շ 0.5 V) and
irreversible further oxidation (E տ 1.0 V). From the electro-
communication point of view redox splitting δE1/2 rather
than absolute peak positions are of prime concern. In the
series under investigation redox splitting is partially re-
Figure 6 Experimental and simulated EPR spectra for the biradi-
cals 3.., 4.., and 6.. (X-band, toluene, 290 K). 3..: giso ϭ 1.986,
a(51V) ϭ 7.28 mT, ΗJEPRΗ ϭ 75 mT (0.070 cmϪ1), line width ϭ
3.3 mT, α ϭ Ϫ0.77, β ϭ Ϫ0.004, γ ϭ 2.8, χ2 ϭ 0.06, α, β, γ are the
coefficients in the line width equation which, in spectral simulations
allow for mI dependence of the line shape [12]. 4..: giso ϭ 1.986,
a(51V) ϭ 7.20 mT, ΗJEPRΗ ϭ 81 mT (0.076 cmϪ1), line width ϭ
3.2 mT, α ϭ Ϫ0.51, β ϭ Ϫ0.037, γ ϭ 0.89, χ2 ϭ 0.083. 6..: giso
ϭ
1.986, a(51V) ϭ 7.20 mT, ΗJEPRΗ ϭ 83 mT (0.077 cmϪ1), line width ϭ
3.1 mT, α ϭ Ϫ0.52, β ϭ Ϫ0.034, γ ϭ 0.92, χ2 ϭ 0.084.
solved only in the cathodic region for which the gradation
δE1/2 (0/Ϫ/2Ϫ) 3.. > 6.. > 4.. is observed. In search for an
electrostructural correlation one notices that 3.. contains a
nearly linear disiloxane bridge with the shortest Si-O bond
lengths in the series and a partial conversion of the oxygen
lone pairs into O-Si πϪbonding electron pairs. Possibly it
is the attendant increase in polarizability of these electrons
which fosters the transmission of an electrostatic effect ac-
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
2391

C. Elschenbroich, F. Lu, O. Burghaus, K. Harms, M. Nowotny
contraction during oxidation tends to separate the charges.
Note that the redox processes at trovacene are highly metal
centered as the HOMO of 1. is almost exclusively of Vdz2
character.
In conclusion, the work described here illustrates that
modifying the nature of a spacer, keeping the number of
spacer atoms constant not necessarily has a pronounced
effect on the extent of the magnetic exchange interaction.
Further studies on bi[5]trovacenyls separated by three-atom
spacers will reveal, whether the near identity of the J values
found for 3.., 4.., and 6.. is accidental or whether the re-
sponse of intramolecular electron-spin exchange coupling
to changes of the spacer structure is smaller than surmised
originally. Electrocommunication, on the other hand, is af-
fected significantly as demonstrated by the redox splittings
derived from cyclic voltammetry.
Experimental Section
All chemical manipulations and physical measurements were per-
formed in an atmosphere of purified dinitrogen or argon in dry
and degassed solvents employing standard Schlenk techniques [16].
Trovacene (η7-C7H7)V(η5-C5H5) (1.) [19] and (ClSiMe2)2CH2 [20]
and were prepared as described in the literature.
X-ray data collection
Single crystals were mounted at the top of a glass capillary in a
drop of inert oil under a flow of cold gaseous nitrogen. The data
were collected with a STOE IPDS2 diffractometer employing Mo
˚
Kα radiation (λ ϭ 0.71073 A). The structures were solved by direct
methods. Full-matrix least-squares on F2 were performed as struc-
ture refinements. All non-hydrogen atoms were refined aniso-
tropically. Crystal data, technical programs, as well as software
were listed in Table 6. The corresponding deposition numbers for
2., 3.., 4.., and 5. are CCDC 646027/646028, 646030, 646031, and
646029, respectively. Copies of data may be obtained free of charge
on application to CCDC, 12 Union Road, Cambridge CB2 1EZ,
UK (http://www.ccdc.cam.ac.uk).
Figure 7 Cyclic voltammetry traces for the complexes 3.., 4.., and
6.. (DME, (n-Bu)4N(ClO4), Ϫ40 °C, v ϭ 100 mV sϪ1, SCE).
Table 5 Electrochemical data from cyclic voltammetry for the
complexes 3.., 4.., and 6...
Spacer
ϪSiMe2CH2SiMe2- ϪSiMe2OSiMe2- ϪSiPh2OSiPh2-
3..
Ϫ2.56
50
4..
Ϫ2.53
50
6..
Ϫ2.54
58
[5]Trovacenyl-diphenylsilanol (2.)
E1/2 (-/2-) / V
∆Ep / mV
E1/2 (0/-) / V
∆Ep / mV
E1/2 (ϩ/0) / V
∆Ep /mV
Ep,a (nϩ/ϩ) / V
Ep,c (ECE) / V Ϫ0.18
δE1/2 (0/-/2-) / mV 82
Ϫ2.48
50
0.25
54
Ϫ2.49
50
0.24
54
1.07
Ϫ0.20
32 (estimated)
Ϫ2.41
56
0.28
58
1.18
Ϫ0.20
130
To a solution of trovacene (1.) (860 mg, 4.15 mmol) in 150 ml Et2O,
n-butyl lithium (2.6 ml, 4.16 mmol, 1.6 M/hexane) was added and
stirred at room temperature overnight to give a red-brown solution.
After cooling to 0 °C, Ph2SiCl2 (0.22 ml, 1.0 mmol) was added
dropwise in 1 h, then the mixture was warmed to r.t. and more
Ph2SiCl2 (0.22 ml, 1.0 mmol) was added. The mixture was stirred
overnight at room temperature, filtered through Celite and freed of
all the volatiles. Purification by chromatography (Al2O3, 2ϫ30 cm,
eluted by EtOH/Et2O) gave 2. as a purple solid. The single crystal
suitable for X-ray diffraction was grown from saturated ethanol
1.08
General conditions: working electrode: glassy carbon; E vs SCE; solvent:
DME; 0.1 m TBAP; T ϭ Ϫ40°.
ross this bridge, resulting in a detectable redox splitting. In
a related fashion the smaller redox splitting observed for 6..
could be traced to CH-Si hyperconjugation. The fact that
redox splitting for reduction exceeds that for oxidation is
plausible because V d-orbital expansion upon reduction
brings the charges into closer proximity whereas d-orbital
solution. EPR: X-band, toluene, 290 K, Aiso(51V) ϭ 72.6 G, giso
ϭ
1.9834. Yield: 220 mg (0.54 mmol, 13 %).
Elemental analysis: C24H22OSiV (405.26): calcd C 71.12, H 5.43;
found C 69.13, H 5.78 %. EI-MS (70 eV) m/z ϭ 405 [Mϩ] (48 %),
207 [TVCϩ] (100 %).
2392
www.zaac.wiley-vch.de
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394

Trovacene Chemistry. 16
Table 6 Crystallographic data of 2., 3.., 4.., and 5..
2 2. EtOH
0.36ϫ0.27ϫ0.21
monoclinic
Cc
2. PE/benzene
0.41ϫ0.30ϫ0.25
monoclinic
P21/c
C24H22OSiV
405.45
10.0867(9)
25.1082(16)
7.9094(8)
105.880(7)
1926.7(3)
4
3.. ·1/2 C6H6
0.40ϫ0.30ϫ0.08
orthorhombic
Ccca
C51H45OSi2V2
831.68
20.0119(10)
34.109(2)
12.1983(6)
90
8326.5(8)
8
4..
5.
crystal size /mm
crystal system
space group
Formula
0.30ϫ0.24ϫ0.12
monoclinic
P21/n
C28H34Si2V2
544.61
7.8240(5)
14.1395(9)
23.7599(16)
94.410(8)
2620.7
0.33ϫ0.27ϫ0.13
triclinic
¯
P1
C50H50O3Si2V2
C36H32O2Si2V
603.74
formula weight
856.96
˚
a /A
18.3952(10)
10.0390(6)
23.7742(13)
108.341(4)
4167.3(4)
4
8.8101(6)
10.3305(8)
17.7825(13)
86.792(6)
1492.01(19)
2
˚
b /A
˚
c /A
β /°
3
˚
volume /A
Z
4
density /mg/m3
absorbtion coefficient
F(000)
1.366
1.398
1.338
1.380
1.344
0.549 mmϪ1
1792
0.588 mmϪ1
844
0.545 mmϪ1
3496
0.824 mmϪ1
1136
0.4400 mmϪ1
630
Θrange
hkl index ranges
1.80 to 26.25°
Ϫ22ՅhՅ22,
Ϫ12ՅkՅ12,
Ϫ29ՅlՅ29
29818
1.62 to 25.38°
Ϫ12ՅhՅ11,
Ϫ30ՅkՅ30,
Ϫ9ՅlՅ9
16288
2.04 to 25.00°
Ϫ23ՅhՅ23,
Ϫ40ՅkՅ40,
Ϫ14ՅlՅ14
29322
1.68 to 25.00°
Ϫ9ՅhՅ9,
Ϫ15ՅkՅ16,
Ϫ28ՅlՅ20
11600
2.14 to 26.28°
Ϫ10ՅhՅ10,
Ϫ12ՅkՅ12,
Ϫ22ՅlՅ22
24390
reflections collected
indepentent reflections
obsd reflns [I>2σ(I)]
reflections used
8339 [R(int) ϭ 0.0372] 3380 [R(int) ϭ 0.0251] 3682 [R(int) ϭ 0.0549] 4565 [R(int) ϭ 0.0345] 5972 [R(int) ϭ 0.0725]
7944
8399
0.199 and Ϫ0.196
715
1.034
0.0229
0.0595
3069
3380
0.296 and Ϫ0.279
332
1.040
0.0274
0.0734
2680
3682
0.934 and Ϫ0.318
353
0.979
0.0400
0.1021
3057
4565
0.310 and Ϫ0.205
368
0.895
0.0326
0.0748
4311
5972
0.437 and -0.413
375
0.935
0.0400
0.1099
Ϫ3
˚
largest diff. Peak/ hole /e.A
Parameters
GOF F2
R1
wR2
2ϫ30 cm, toluene/PE and THF as eluents) to afford 4.. as violet-
blue solid. Yield: 290 mg (0.53 mmol, 29 %).
Di([5]trovacenyl)tetraphenyl-disiloxane (3..) and
([5]trovacenyl)diphenylsiloxy-diphenylsilanol (5.)
Elemental analysis: C28H34OSi2V2 (544.335): calcd C 61.78, H 6.25;
found C 62.56, H 6.39 %. EI-MS (70 eV) m/z ϭ 544 [Mϩ] (36 %);
452 [Mϩ-C7H7] (26 %); 91 [C7H7ϩ] (100 %).
To a solution of trovacene (1.) (983 mg, 4 mmol) in Et2O (150 ml),
n-butyl lithium (3.0 ml, 1.6 M/hexane, 4.8 mmol) was added and
the mixture was stirred at room temperature overnight. Then 1,3-
dichlorotetraphenyldisiloxane (0.45 ml, 1.19 mmol) in 20 ml Et2O
was added in 2 h with refluxing. After addition, the mixture was
refluxed overnight with vigorous stirring. The suspension was fil-
tered through Celite, the filtrate was evaporated under reduced
pressure. Isolation by chromatography (Al2O3, 2.5ϫ30 cm) gave 3..
from a blue band eluted with toluene/Et2O (3:1), followed by a pale
blue band (EtOH: Et2O ϭ 1:1) which contains 5.. The single crys-
tals suitable for X-ray diffraction were grown from benzene/PE at
8 °C. 3..: Yield: 314 mg (0.40 mmol, 20 %).
Di([5]trovacenyldimethylsilyl)methane (6..)
(ClSiMe2)2CH2 (0.16 ml, 0.8 mmol) was added dropwise to a solu-
tion of lithiotrovacene ( from trovacene (1.) 670 mg, 3.24 mmol, n-
butyl lithium 2 ml, 3.2 mmol, 1.6 M/hexane) at room temperature
with vigorous stirring. After stirring overnight, the suspension was
filtered through Celite, and the filtrate was evaporated in vacuum,
the residue was purified by chromatography (Al2O3) to give 6.. as
pale blue solid. Yield: 140 mg (0.26 mmol 16 %).
Elemental analysis: C48H42OSi2V2 (792.651): calcd C 72.73, H 5.30;
found C 71.79, H 5.27 %. EI-MS (70 eV) m/z ϭ 792 [Mϩ] (90 %),
Elemental analysis: C29H36Si2V2 (542.346): calcd C, 64.22; H, 6.69;
found: C, 63.58; H, 7.26 %. EI-MS (70 eV) m/z ϭ 542 [Mϩ]
(100 %); 451 [Mϩ-C7H7] (50 %); 264 [TVC-SiMe2ϩ] (4 %); 142
[C5H4VSiϩ] (87 %); 116 [C5H4Vϩ] (18 %).
ϩ
701 [Mϩ-C7H7] (27 %), 142 [C5H4VSiϩ] (25 %), 28 [C2H4
(100 %).
]
5.: Yield : 130 mg (0.22 mmol, 16 %).
Acknowledgement. This work was supported by the Deutsche
Forschungsgemeinschaft and the Fonds der Chemischen Industrie.
Elemental analysis: C36H32O2Si2V (603.506): calcd C 71.64, H 5.30;
found C 70.43, H 5.22 %. EI-MS (70 eV) m/z ϭ 603 [Mϩ] (100 %),
207 [TVCϩ] (20 %).
References and Notes
Di([5]trovacenyl)tetramethyl-disiloxane (4..)
[1] Part 15: Ch. Elschenbroich, F. Lu, O. Burghaus, C. Pietzonka,
K. Harms, Chem. Comm. 2007, 3201.
[2] (a) V. Chandrasekhar, Inorganic and Organometallic Polymers,
To a red-brown solution of lithiotrovacene (from trovacene (1.)
760 mg, 3.67 mmol and n-butyl lithium 2.3 ml, 3.68 mmol, 1.6 M/
hexane) in Et2O (150 ml), ClMe2SiOSiMe2Cl (0.18 ml, 0.92 mmol)
was added dropwise in 2 h and the mixture was refluxed for 3 h.
The suspension was filtered through Celite and freed of all the vol-
atiles. Purification was achieved by chromatography (Al2O3,
´
Springer, Berlin 2005; (b) P. Nguyen, P. Gomez-Elipe, I. Man-
ners, Chem. Rev. 1999, 99, 1515; (c) R. J. Puddephatt, Chem.
Commun. 1998, 1055; (d) I. Manners, Science 2001, 294, 1664;
(e) I. Manners, Synthetic Metal Containing Polymers, Wiley-
VCH, Weinheim (2003); (f) G. R. Newkome, E. He, C. N.
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
2393

C. Elschenbroich, F. Lu, O. Burghaus, K. Harms, M. Nowotny
Moorefield, Chem. Rev. 1999, 99, 1689; R. D. Archer, Inor-
ganic and Organometallic Polymers, Wiley-VCH, Weinheim
(2001); (g) M. S. Khan, M. R. A. Al-Mandhary, M. K. Al-
Suti, A. K. Hisahm, P. R. Raithby, B. Ahrens, M. F. Mahon,
L. Male, E. A. Marseglia, E. Tedesco, R. H. Friend, A. Köhler,
N. Feeder, S. J. Teat, J. Chem. Soc. Dalton Trans. 2002, 1358;
(h) T. Yamamoto, K.-I. Sanechika, A. Yamamoto, M. Katada,
I. Motoyama, H. Sano, Inorg. Chim. Acta 1983, 73, 75; (i) W.
J. Patterson, S. P. McManus, C. U. Pittman, Jr., J. Polym. Sci.
Polym. Chem. 1974, 12, 837; (j) T. Itoh, H. Saito, S. Iwatsuki,
J. Polym. Sci. Polym. Chem. 1995, 33, 1589; (k) I. Manners,
Comprehensive Organometallic Chemistry. III (R. M.:
Crabtree, D. M. P. Mingos, Eds) 2007, 12, 295.
[9] K. L. Malisza, L. C. F. Chao, J. F. Britten, B. G. Sayer, G.
Jaouen, S. Top, A. Decken, M. J. McGlinchey, Organometallics
1993, 12, 2462.
[10] C. Glidewell, D. C. Liles, J. Chem. Soc. Chem. Commun.
1977, 632.
[11] F. Cervantes-Lee, H. K. Sharma, K. H. Pannell, A. Derecskei-
Kovacs, D. S. Marynick, Organometallics 1998, 17, 3701.
[12] (a) C. Elschenbroich, M. Wolf, O. Schiemann, K. Harms,
O. Burghaus, J. Pebler, Organometallics 2002, 21, 5810;
(b) O. Kahn, Molecular Magnetism, VCH, Weinheim, 1993,
Chapter 7.
[13] F. E. Mabbs, Chem. Soc. Rev. 1993, 22, 313.
[14] C. Elschenbroich, J. Plackmeyer, M. Nowotny, A. Behrendt,
K. Harms, J. Pebler, O. Burghaus, Chem. Eur. J. 2005, 11, 7427.
[15] (a) W. Skibar, H. Kopacka, K. Wurst, C. Salzmann, K.-H.
Ongania, F. F. de Biani, P. Zanello, B. Bildstein, Organometall-
ics 2004, 23, 1024; (b) B. Bildstein, O. Loza, Y. Chizhov, Or-
ganometallics 2004, 23, 1825; (c) B. Bildstein, Coord. Chem.
Rev. 2000, 206-207, 369. Note: [n-1] Cumulenes are defined as
chains of n carbon atoms connected by (n-1) double bonds.
[16] C. Elschenbroich, J. Plackmeyer, M. Nowotny, K. Harms, J.
Pebler, O. Burghaus, Inorg. Chem. 2005, 44, 955.
[17] S. Shambayati, S. L. Schreiber, J. F. Blake, S. G. Wierschke,
W. L. Jorgensen, J. Am. Chem. Soc. 1990, 112, 697.
[18] J. R. Durig, M. J. Flanagan, V. F. Kalasinsky, J. Chem. Phys.
1977, 66, 2775.
[19] (a) R. B. King, F. G. A. Stone, J. Am. Chem. Soc. 1959, 81,
5263; (b) C. Floriani, J. Chem. Soc. Dalton Trans. 1976, 1046.
[20] (a) U. Herzog, G. Rheinwald, J. Organomet. Chem. 2001, 628,
[3] The number in brackets indicates the site of peripheral substi-
tution: [5]trovacenyl derivatives are functionalized at the cyclo-
pentadienyl ligand and [7]trovacenyl derivatives at the cyclo-
heptatrienyl ligand.
[4] See ref. 1 and previous papers in the series Trovacene Chemis-
try.
[5] (a) H. Puff, K. Braun, H. Reuter, J. Organomet. Chem. 1991,
409, 119; (b) S. A. Bourne, L. R. Nassimbeni, K. Skobridis,
E. Weber, J. Chem. Soc. Chem. Commun. 1991, 282; (c) E. A.
Babaian, M. Huff, F. A. Tibbals, D. C. Hrncir, J. Chem. Soc.
Chem. Commun. 1990, 306.
[6] P. D. Lickiss, Adv. Inorg. Chem. 1995, 42, 147.
[7] (a) G. A. Jeffrey, An Introduction to Hydrogen Bonding, Oxford
University Press, New York P. 12, 1997; (b) J. E. Huheey, Inor-
ganic Chemistry, 3rd Ed, Harper, Cambridge 1983.
[8] (a) D. Braga, F. Grepioni, E. Tedesco, Organometallics 1998,
17, 2669; (b) S. Harder, Chem. Eur. J. 1999, 5, 1852.
´
´
133; (b) G. Delris, M. Birot, J. Dunogues, B. Barbe, M. Pet-
taud, M. Lefort, J. Organomet. Chem. 1984, 266, 1.
2394
www.zaac.wiley-vch.de
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 2386Ϫ2394