Facile cleavage of Si-Si or Si-Ge bonds in the reactions of disilanes or germylsilanes with cobalt carbonyl

Article abstract of DOI:10.1016/S0022-328X(98)01016-X

The disilane (Ph₂HSi)₂ reacted at room temperature with Co₂(CO)₈ to yield Co(SiPh₃)(CO)₄ as a major product. Corresponding monosilyl derivatives were formed also when other disilanes Ph₃SiSiR₂H (R₂=Ph₂, PhMe, Me₂, Et₂), Me₃SiSiPh₂H and PhMe₂SiSiPhMeH were reacted with Co₂(CO)₈ under the same conditions. A mechanism based on silyl-silylene intermediates is proposed. The germylsilanes Ph₃GeSiR₂H (R=Ph, Et) with Co₂(CO)₈ also gave Co(SiPh₃)(CO)₄ by selective elimination of:GePh₂ which was trapped by Co₂(CO)₈ as Ph₂Ge[Co(CO)₄]₂. Crystal structures of Co(SiPh₃)(CO)₄ and Ph₂Ge[Co(CO)₄]₂ are presented.

Full text of DOI:10.1016/S0022-328X(98)01016-X

Journal of Organometallic Chemistry 577 (1999) 181–188
Facile cleavage of Si–Si or Si–Ge bonds in the reactions of disilanes
or germylsilanes with cobalt carbonyl
J. Scott McIndoe, Brian K. Nicholson *
Department of Chemistry, Uni6ersity of Waikato, Pri6ate Bag 3105, Hamilton, New Zealand
Received 10 August 1998
Abstract
The disilane (Ph2HSi)2 reacted at room temperature with Co2(CO)8 to yield Co(SiPh3)(CO)4 as a major product. Corresponding
monosilyl derivatives were formed also when other disilanes Ph3SiSiR2H (R2=Ph2, PhMe, Me2, Et2), Me3SiSiPh2H and
PhMe2SiSiPhMeH were reacted with Co2(CO)8 under the same conditions. A mechanism based on silyl–silylene intermediates is
proposed. The germylsilanes Ph3GeSiR2H (R=Ph, Et) with Co2(CO)8 also gave Co(SiPh3)(CO)4 by selective elimination of:GePh2
which was trapped by Co2(CO)8 as Ph2Ge[Co(CO)4]2. Crystal structures of Co(SiPh3)(CO)4 and Ph2Ge[Co(CO)4]2 are presented.
© 1999 Elsevier Science S.A. All rights reserved.
Keywords: Cobalt; Carbonyl; Triphenylsilyl complexes; Silylene; Germylene
base [8], trapping of the expelled silylene [9], and the
recent isolation of a true intermediate in the photo-
chemical silylene-elimination/isomerisation process [10].
A stable stannyl/stannylene complex of manganese has
also been characterised fully [11]. This mechanism has
been adopted for other systems including closely-related
indenyl complexes Fe(p5-C9H7)(CO)2 [12], the
analogous Ru(p5-C5H5)(CO)2 examples, and polysilyl
complexes involving tungsten [13], palladium and plat-
inum [14], rhodium [15], iridium [15], and nickel [16].
For most of these systems relatively vigorous condi-
tions, usually UV-irradiation, has been necessary to
effect the elimination reactions.
So far no examples of similar behaviour have been
clearly defined for cobalt complexes. Kerber and Pak-
kanen reacted (Me2HSi)2 with Co2(CO)8 and isolated
(v2-Me2Si)2Co2(CO)6 in 15% yield [17]. A variety of
other products were seen in the reaction but could not
be identified positively [17]. A mechanism was proposed
based on the intermediacy of complexes with terminal
silylene ligands. One difficulty is that methyl-substituted
silicon-cobalt complexes are known to be thermally
labile and air-sensitive; Co(SiMe3)(CO)4 is best handled
1. Introduction
Since Pannell and co-workers reported the de-
oligomerisation of Fe(SiMe2SiMe3)(p5-C5H5)(CO)2 to
Fe(SiMe3)(p5-C5H5)(CO)2 in 1974 [1], there has been
sustained interest in the chemistry of the silicon–silicon
bond in transition metal-substituted oligosilanes [2].
The mechanism of the photodeoligomerisation was pro-
posed [3] to occur via a series of equilibrating silyl(si-
lylene) intermediates formed upon photoelimination of
a CO ligand followed by a 1,2-silyl shift; a series of
1,3-migrations led to scrambling of the silicon sub-
stituents. Finally (photo)expulsion of the silylene frag-
ment gave the mono-silyl product. Subsequent work,
mainly from the laboratories of Pannell and Ogino, has
supported this general mechanism and has included
labelling [4], mixing of silicon substituents [5], direct
observation of the intermediate at low temperature [6],
intramolecular stabilisation of bis(silylene) complexes
[7], trapping of the silyl(silylene) intermediate using a
* Corresponding author. Fax: +64-7-8384219.
E-mail address: b.nicholson@waikato.ac.nz (B.K. Nicholson)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(98)01016-X

182
J.S. McIndoe, B.K. Nicholson / Journal of Organometallic Chemistry 577 (1999) 181–188
on a high-vacuum line [18], while [Me2{(OC)4Co}Si]2O
was described by Greene and Curtis [19] as extremely
thermolabile and it decomposed slowly even at −78°C
under a nitrogen atmosphere. However, corresponding
phenyl species such as Co(SiPh3)(CO)4 are more
amenable to study, although even these do not survive
normal chromatographic separation.
[20]. The disilane starting materials were from treat-
ment of R2SiHCl with R3SiLi, adapting published
methods [21]. Me3SiPh2SiH was prepared by acid hy-
drolysis of Me3SiPh2SiLi. Several literature routes to
(Ph2HSi)2 have been published [22], but we used rapid
stirring of Ph2SiHCl with lithium metal in THF for
three days.
This present paper describes some reactions of
phenyl-substituted
polysilane
compounds
with
2.2. Instrumentation.
Co2(CO)8 under mild conditions. Mixed germyl-silyl
substrates were investigated also.
Infrared spectra were recorded on a Digilab FTS-40
FTIR spectrophotometer. NMR spectroscopy was per-
formed using a Bruker AC300P Multinuclear FT spec-
trometer. Elemental analysis was performed by the
Campbell Microanalytical Laboratory, University of
Otago. Melting points were measured on a Reichart
Thermopan melting point apparatus and are
uncorrected.
2. Experimental
2.1. General
All manipulations were carried out in an oxygen-free
N2 atmosphere with rigorously dried solvents. Reac-
tions were conducted either in Schlenk flasks or in
sealed glass ampoules. Starting silicon compounds
Ph3SiCl, Ph2SiHCl and PhMeSiHCl were purchased
from Petrarch and used as supplied. Co2(CO)8 was
freshly sublimed before use. Me2SiHCl was prepared
using a published method, while Et2SiHCl was pre-
2.3. Reactions
2.3.1. Reaction of (Ph2HSi)2 with Co2(CO)8
To a Schlenk flask was added (Ph2HSi)2 (0.200 g,
0.55 mmol), Co2(CO)8 (0.187 g, 0.55 mmol) and
petroleum spirits (10 ml), and the mixture was left to
stir for 2 days. An IR spectrum of the crude mixture
showed the presence of Co2(CO)8 (ꢀ40%), Co4(CO)12
(ꢀ10%), CoH(CO)4 (ꢀ5%) and the remainder
Co(SiPh3)(CO)4 with peaks at 2095 (m), 2034 (m) and
2006 (vs) cm−1. Prolonged pumping under high vac-
uum removed the CoH(CO)4 and the remaining species
were separated by fractional crystallisation. Transpar-
ent crystals of Co(SiPh3)(CO)4 (32 mg, 0.086 mmol,
16%) were hand separated in a glove box from black
Co4(CO)12 and orange Co2(CO)8. Data for
Co(SiPh3)(CO)4: m.p. 158–160°C. IR w(CO):
(petroleum spirits, cm−1) 2095 (m), 2034 (m), 2006 (vs);
(CH2Cl2, cm−1) 2096 (m), 2036 (m), 2002 (vs). NMR
i
pared similarly to the published synthesis of Pr2SiHCl
Table 1
Atomic coordinates and equivalent isotropic displacement parameters
for Ph3SiCo(CO)4
x
y
z
Ueq
Co(1)
Si(1)
O(1)
O(2)
O(3)
O(4)
C(1)
C(2)
C(3)
0.7339(1)
0.6254(1)
0.8707(2)
0.6803(2)
0.4892(2)
0.9943(2)
0.8180(3)
0.6989(3)
0.5852(3)
0.8913(3)
0.6826(2)
0.8252(3)
0.8643(3)
0.7604(3)
0.6177(3)
0.5794(3)
0.4202(2)
0.3392(3)
0.1880(3)
0.1146(3)
0.1914(3)
0.3430(3)
0.6859(2)
0.7432(3)
0.7915(3)
0.7828(3)
0.7263(4)
0.6776(3)
0.6381(1)
0.7929(1)
0.4442(2)
0.8533(2)
0.4956(2)
0.6289(2)
0.5182(3)
0.7713(3)
0.5522(2)
0.6338(2)
0.9591(2)
0.9798(3)
1.1040(3)
1.2119(3)
1.1950(3)
1.0704(2)
0.8111(2)
0.8039(2)
0.8233(3)
0.8508(3)
0.8584(3)
0.8384(3)
0.7381(2)
0.8245(3)
0.7864(3)
0.6608(4)
0.5734(3)
0.6111(3)
0.8122(1)
0.6875(1)
0.9626(2)
0.9936(2)
0.8409(2)
0.5696(2)
0.9044(3)
0.9223(3)
0.8265(2)
0.6625(3)
0.6860(2)
0.6620(3)
0.6586(3)
0.6795(3)
0.7028(3)
0.7064(3)
0.7600(2)
0.6778(3)
0.7283(4)
0.8606(4)
0.9439(3)
0.8946(3)
0.5102(2)
0.4040(2)
0.2722(3)
0.2444(3)
0.3459(3)
0.4787(3)
0.034(1)
0.030(1)
0.069(1)
0.063(1)
0.054(1)
0.055(1)
0.046(1)
0.042(1)
0.040(1)
0.040(1)
0.031(1)
0.051(1)
0.058(1)
0.052(1)
0.056(1)
0.044(1)
0.034(1)
0.044(1)
0.061(1)
0.063(1)
0.057(1)
0.047(1)
0.035(1)
0.043(1)
0.059(1)
0.064(1)
0.065(1)
0.052(1)
1
(CDCl3) H: l 7.58, 7.38 (m, C6H5); 13C: l 198.8 (CO),
138.3 (ipso), 135.5 (ortho), 129.6 (para), 128.0 (meta)
(C6H5).
C(4)
C(11)
C(12)
C(13)
C(14)
C(15)
C(16)
C(21)
C(22)
C(23)
C(24)
C(25)
C(26)
C(31)
C(32)
C(33)
C(34)
C(35)
C(36)
2.3.2. Reaction of Ph3SiH with Co2(CO)8
Ph3SiH (0.177 g, 0.680 mmol), Co2(CO)8 (0.166 g,
0.340 mmol) and petroleum spirits (10 ml) were added
to a Schlenk flask and allowed to react overnight. The
pale brown solution was cooled to −20°C to give
white crystals of Co(SiPh3)(CO)4 (0.153 g, 0.412 mmol,
61%), spectroscopically identical to that synthesised
indirectly.
2.3.3. Reaction of Ph3SiSiPh2H with Co2(CO)8
The reaction between Ph3SiSiPh2H (0.100 g, 0.226
mmol) and Co2(CO)8 (0.039 g, 0.113 mmol) in
petroleum spirits (10 ml) proceeded slowly over 3 days.
After this time, the main carbonyl-containing com-

J.S. McIndoe, B.K. Nicholson / Journal of Organometallic Chemistry 577 (1999) 181–188
183
Table 2
pounds were Co2(CO)8 (ꢀ50%), Co4(CO)12 (ꢀ5%)
and Co(SiPh3)(CO)4 (ꢀ40%).
Atomic coordinates and equivalent isotropic displacement parameters
for Ph2Ge[Co(CO)4]2
2.3.4. Reaction of Ph3SiSiPhMeH with Co2(CO)8
x
y
z
Ueq
Ph3SiSiPhMeH (0.262 g, 0.700 mmol), Co2(CO)8
(0.118 g, 0.345 mmol) and petroleum spirits (ꢀ1 ml)
were sealed in a small ampoule and left to react in the
dark. After 2 months, crystals of Co(SiPh3)(CO)4 (15
mg, 0.035 mmol, 5%) had precipitated. The solution IR
showed mainly Co(SiPh3)(CO)4, with small (ꢀ10%)
amounts of Co2(CO)8 present. A reaction in a sealed
NMR tube between Co2(CO)8 (31 mg, 0.091 mmol) and
Ph3SiSiPhMeH (69 mg, 0.18 mmol) in C6D6 also
showed mainly Co(SiPh3)(CO)4 as product (by 13C-
NMR), with a complex methyl region (three peaks in
Ge(1)
Co(1)
Co(2)
Ge(2)
Co(3)
Co(4)
C(1)
C(2)
C(3)
C(4)
C(5)
C(6)
C(7)
C(8)
C(9)
C(10)
C(11)
C(12)
C(13)
C(14)
C(15)
C(16)
O(1)
0.1284(1)
0.2282(1)
0.0340(1)
0.1196(1)
0.1073(1)
0.1384(1)
0.3011(2)
0.2606(2)
0.2127(2)
0.1949(2)
0.1552(1)
0.2312(1)
0.2425(1)
0.6656(1)
0.5786(1)
0.5893(1)
0.2868(2)
0.1368(2)
0.2342(2)
0.3057(2)
0.3075(2)
0.1427(2)
0.2964(2)
0.2651(2)
0.5161(2)
0.6755(3)
0.5148(2)
0.5626(3)
0.5364(2)
0.5826(2)
0.5165(2)
0.6848(3)
0.3219(2)
0.0790(2)
0.2362(2)
0.3557(2)
0.3495(2)
0.0807(1)
0.3335(2)
0.2794(2)
0.4755(2)
0.7354(2)
0.4709(2)
0.5504(2)
0.5057(2)
0.5775(2)
0.4681(2)
0.7429(2)
0.0767(2)
0.0820(2)
0.0219(2)
−0.0442(3)
−0.0505(2)
0.0097(2)
0.0833(2)
0.1067(2)
0.0530(3)
−0.0222(3)
−0.0452(3)
0.0069(2)
0.7355(2)
0.7138(2)
0.7633(3)
0.8352(3)
0.8583(2)
0.8092(2)
0.7422(2)
0.8108(2)
0.8612(2)
0.8436(2)
0.7767(3)
0.7264(2)
0.4108(1)
0.4205(1)
0.4029(1)
0.4183(1)
0.5794(1)
0.2505(1)
0.4265(3)
0.3676(3)
0.5612(3)
0.3306(3)
0.4068(3)
0.4145(3)
0.5113(3)
0.2651(3)
0.6984(3)
0.6457(3)
0.5263(3)
0.5372(3)
0.1260(3)
0.2963(3)
0.2953(3)
0.1890(3)
0.4277(3)
0.3331(2)
0.6503(2)
0.2732(3)
0.4111(3)
0.4189(3)
0.5811(3)
0.1784(2)
0.7715(2)
0.6928(3)
0.4978(3)
0.5158(3)
0.0472(3)
0.3207(3)
0.3191(3)
0.1441(2)
0.5308(3)
0.6141(3)
0.6929(3)
0.6908(3)
0.6094(3)
0.5306(3)
0.2838(3)
0.1897(3)
0.1022(3)
0.1090(4)
0.2003(4)
0.2888(3)
0.4068(3)
0.3444(3)
0.3430(3)
0.4028(4)
0.4650(3)
0.4670(3)
0.4311(3)
0.3625(3)
0.3579(3)
0.4214(4)
0.4912(4)
0.4958(3)
0.024(1)
0.030(1)
0.030(1)
0.027(1)
0.033(1)
0.033(1)
0.039(1)
0.036(1)
0.041(1)
0.040(1)
0.038(1)
0.035(1)
0.040(1)
0.040(1)
0.041(1)
0.044(1)
0.042(1)
0.047(1)
0.047(1)
0.043(1)
0.041(1)
0.046(1)
0.057(1)
0.049(1)
0.064(1)
0.061(1)
0.056(1)
0.047(1)
0.060(1)
0.060(1)
0.058(1)
0.060(1)
0.061(1)
0.072(1)
0.074(1)
0.063(1)
0.061(1)
0.070(1)
0.029(1)
0.035(1)
0.043(1)
0.044(1)
0.042(1)
0.036(1)
0.031(1)
0.041(1)
0.057(1)
0.063(2)
0.056(1)
0.040(1)
0.032(1)
0.041(1)
0.052(1)
0.056(1)
0.051(1)
0.039(1)
0.032(1)
0.039(1)
0.049(1)
0.053(1)
0.051(1)
0.040(1)
−0.0351(2)
−0.0040(1)
0.0690(2)
0.0532(2)
0.1000(2)
0.1219(2)
0.1680(2)
0.0284(2)
0.1537(2)
0.2178(2)
0.0814(2)
0.1142(2)
0.3468(1)
0.2841(1)
0.2061(2)
0.1784(2)
−0.0770(1)
−0.0294(1)
0.0878(1)
0.0613(2)
0.0958(2)
0.1311(2)
0.2064(2)
−0.0222(2)
0.1636(2)
0.2684(1)
0.0454(2)
0.0986(2)
0.1250(2)
0.0828(2)
0.0811(2)
0.1222(2)
0.1650(2)
0.1663(2)
0.1277(2)
0.1569(2)
0.1555(2)
0.1242(2)
0.0944(2)
0.0962(2)
0.0450(1)
−0.0061(2)
−0.0589(2)
−0.0612(2)
−0.0117(2)
0.0416(2)
0.1907(2)
0.1929(2)
0.2456(2)
0.2965(2)
0.2946(2)
0.2417(2)
1
the H SiMe region, at l 1.08, 1.03 and 0.56).
2.3.5. Reaction of Ph3SiSiMe2H with Co2(CO)8
Ph3SiSiMe2H (16.9 mg, 0.0531 mmol) and Co2(CO)8
(9.0 mg, 0.026 mmol) in petroleum spirits (5 ml) were
stirred in a Schlenk flask overnight. The w(CO) IR
spectrum showed some Co2(CO)8 (ꢀ20%) and broad
peaks at 2094 (m), 2032 (m) and 2002 (vs) cm−1
(ꢀ80%). Attempted purification of this solution was
unsuccessful. An NMR study of the reaction between
Co2(CO)8 (22 mg, 0.064 mmol) and Ph3SiSiMe2H (43
mg, 0.13 mmol) in C6D6 showed a mixture of products,
with at least five peaks in the 1H SiMe region, at l 1.08,
1.03, 0.76, 0.51 and 0.29.
O(2)
O(3)
O(4)
O(5)
O(6)
O(7)
O(8)
O(9)
O(10)
O(11)
O(12)
O(13)
O(14)
O(15)
O(16)
C(21)
C(22)
C(23)
C(24)
C(25)
C(26)
C(31)
C(32)
C(33)
C(34)
C(35)
C(36)
C(41)
C(42)
C(43)
C(44)
C(45)
C(46)
C(51)
C(52)
C(53)
C(54)
C(55)
C(56)
2.3.6. Reaction of Ph3SiSiEt2H with Co2(CO)8
Ph3SiSiEt2H (0.343 g, 0.991 mmol) and Co2(CO)8
(0.170 g, 0.497 mmol) in petroleum spirits (10 ml) were
stirred in a Schlenk flask overnight. The w(CO) IR
spectrum showed some Co2(CO)8 (ꢀ20%) and broad
peaks at 2093 (m), 2031 (m) and 1999 (vs) cm−1
(ꢀ80%). Attempted purification of this solution was
unsuccessful.
2.3.7. Reaction of Me3SiSiPh2H with Co2(CO)8
Co2(CO)8 (0.227 g, 0.663 mmol) was added to a
Schlenk flask containing Me3SiSiPh2H (ꢀ1 ml, excess)
in petroleum spirits (ꢀ10 ml) and the mixture allowed
to stir overnight. The w(CO) IR spectrum of the result-
ing solution showed peaks due to a small amount of
Co2(CO)8 (ꢀ5%) and peaks at 2091 (m), 2030 (m),
2001 (s) and 1997 (s) cm−1. Attempted purification of
this solution was unsuccessful. An NMR-scale study of
the same reaction showed a complex mixture of prod-
1
ucts, with at least eight peaks in the H SiMe region, at
l 1.08, 1.03, 0.76, 0.63, 0.50, 0.28, 0.21 and 0.13.
2.3.8. Reaction of PhMe2SiSiPhMeH with Co2(CO)8
PhMe2SiSiPhMeH (0.309 g, 1.21 mmol), Co2(CO)8
(0.207 g, 0.605 mmol) and petroleum spirits (ꢀ10 ml)
were added to a Schlenk flask and allowed to stir

184
J.S. McIndoe, B.K. Nicholson / Journal of Organometallic Chemistry 577 (1999) 181–188
Fig. 1. Two views of the structure of Co(SiPh3)(CO)4, a general view and (inset) a view along the Co–Si vector. Bond parameters include:
˚
Co(1)–Si(1) 2.3810(7), Co(1)–C(1) 1.813(3), Co(1)–Ceq (av) 1.787(3), Si(1)–C (av) 1.880(2) A; Co(1)–Si(1)–C (av) 110.65°, Si(1)–Co(1)–C(1)
178.91(8), Si(1)–Co(1)–Ceq (av) 83.86(8)°.
overnight. The w(CO) IR spectrum of the resulting clear
orange/brown solution showed peaks due to Co2(CO)8
(ꢀ20%) and broad peaks at 2092 (m), 2030 (m) and
1999 (vs) cm−1 indicating a mixture of products. At-
tempted purification of this solution was unsuccessful.
bright yellow solution, some small black cubic crystals
(Co4(CO)12) and several large, pale yellow prisms,
shown spectroscopically and by an X-ray crystal struc-
ture analysis to be Ph2Ge[Co(CO)4]2 (19 mg, 0.033
mmol, 18%). The solution showed peaks at 2091 (m),
2030 (m), 2002 (vs) and 1995 (s) cm−1, consistent with
PhEt2SiCo(CO)4. Data for Ph2Ge[Co(CO)4]2: mp 140
°C (dec.). IR w(CO): (petroleum spirits, cm−1) 2100
(w), 2084 (s), 2033 (m), 2023 (s), 2015 (s), 2000 (m).
2.3.9. Reaction of Ph3GeSiPh2H with Co2(CO)8
Ph3GeSiPh2H (0.100 g, 0.206 mmol) was added to a
Schlenk flask containing Co2(CO)8 (35 mg, 0.10 mmol)
and toluene (10 ml). The reaction was stirred overnight.
The w(CO) IR spectrum of the resulting clear yellow
solution showed peaks at 2091 (m), 2033 (m) and 2002
(vs) cm−1 as well as peaks due to CoH(CO)4 at 2054
and 2030 cm-1, and a large number of smaller peaks.
The solution was reduced in volume to ꢀ2 ml,
petroleum spirits (ꢀ5 ml) added and left at −20°C to
crystallise. A pale yellow powder was produced, which
could be partially dissolved in petroleum spirits to give
a spectrum identical to that of Co(SiPh3)(CO)4. The
remaining yellow powder dissolved in toluene to give a
spectrum with peaks at 2100 (w), 2084 (s), 2033 (m),
2023 (s), 2015 (s) and 2000 (m) cm−1. This species was
later identified as Ph2Ge[Co(CO)4]2. No evidence for
Co(GePh3)(CO)4 species was seen.
2.4. X-ray crystallography
Unit cell parameters and intensity data were collected
using a Siemens SMART CCD diffractometer, using
standard collection procedures, with monochromatic
˚
Mo–Kh X-rays (0.71073 A) at 203 K. Corrections for
absorption and other effects was carried out with SAD-
ABS [23], all other calculations used the SHELX97 pro-
grams [24]. The structures were solved by direct
methods, and developed routinely with refinement
based on F2. All non-hydrogen atoms were assigned
anisotropic temperature factors, and hydrogen atoms
were included in calculated positions. Refined coordi-
nates are given in Tables 1 and 2, while selected bond
parameters are in the captions to Figs. 1 and 2.
2.3.10. Reaction of Ph3GeSiEt2H with Co2(CO)8
Ph3GeSiEt2H (0.214 g, 0.548 mmol), Co2(CO)8 (94
mg, 0.27 mmol) and petroleum spirits (ꢀ0.5 ml) were
sealed in a mini-ampoule and left to react in the dark
for 3 weeks. After this time, the ampoule contained a
2.4.1. Crystal data for Co(SiPh3)(CO)4
C22H15CoO4Si, Mr 430.38, triclinic, P1, a=9.8742(5),
˚
b=10.3120(5), c=10.6963(5) A, h=85.687(1), i=
3
˚
69.236(1), k=81.629(1)°, V=1007.21(8) A , Dcalc.
=

J.S. McIndoe, B.K. Nicholson / Journal of Organometallic Chemistry 577 (1999) 181–188
185
Fig. 2. The structure of one of the independent molecules of Ph2Ge[Co(CO)4]2 in the crystal. Bond parameters include: Ge(1)–Co(1) 2.4708(5),
˚
Ge(1)–Co(2) 2.4674(5), Ge(2)–Co(4) 2.4742(6), Ge(2)–Co(3) 2.4749(6), Ge–C (av) 1.967(4), Co–Cax (av) 1.810(4), Co–Ceq (av) 1.790 A;
Co(1)–Ge(1)–Ge(2) 116.06(2), Co(3)–Ge(2)–Co(4) 116.18, C(21)–Ge(1)–C(31) 104.5(1), C(41)–Ge(2)–C(51) 106.7(1), Ge–Co–Cax (av) 178.1(2),
Ge–Co–Ceq (av) 85.3(2)°.
1.420 g cm−3, Z=2, F(000)=440, v(Mo–Kh) 0.921
mm−1, Tmax 0.9272, Tmin 0.7104, crystal size 0.41×
0.14×0.11 mm3.
A total of 9726 reflections, 4444 unique (Rint 0.0263)
was collected 2°BqB28°. Final R1 0.0391 (3451 data
with I\2|(I)), 0.0614 (all data), wR2 0.0866, goodness-
of-fit 1.073, final De +0.331/−0.257.
determination, and by comparison with the product
from Ph3SiH with Co2(CO)8, a known reaction [25].
Although IR evidence suggests that Co(SiPh3)(CO)4
was the major product in the crude reaction mixture,
the isolated yield was not high because of the difficulty
of separating it from the byproducts formed by the
presumed PhSiH species eliminated. This unexpected
rearrangement can be explained by analogy with the
previous photochemically induced processes for the
Fe(p5-C5H5)(CO)2 silyl complexes, as outlined in
Scheme 1. The initial reaction of the Si–H bond with
Co2(CO)8 in the usual manner [25] by elimination of H2
[or HCo(CO)4] and formation of a Si–Co bond will
give HPh2SiSiPh2Co(CO)4. Loss of CO from the cobalt
centre is then followed by a 1,2-silyl migration to give a
silyl/silylene intermediate which undergoes redistribu-
tion of the substituents on the silicon atoms by a series
of 1,3-shifts until the silylene fragment is replaced by
CO to generate the stable complex Co(SiPh3)(CO)4.
The second Si–H bond in the starting silane may also
be converted to a Si–Co(CO)4 group before rearrange-
ment, but in this case some (v2-Ph2Si)2Co2(CO)6 might
have been expected to be formed, but no CO vibrations
assignable to this species were observed in infrared
spectra. The difference between the (R2HSi)2 systems
(R=Me, Ph) is presumably the relative rates of the
Si–H/Co2(CO)8 reaction and the silyl/silylene rear-
rangement process for the two cases.
2.4.2. Crystal data for Ph2Ge[Co(CO)4]2
C20H10Co2GeO8, Mr 568.73, orthorhomic, Pna21,
˚
a=21.5209(2), b=15.9964(2), c=12.5992(2) A, V=
4337.4(1) A , Dcalc. =1.742 g cm−3, Z=8, F(000)=
3
˚
2240, v(Mo–Kh) 2.932 mm−1, Tmax 0.4967, Tmin
0.3877, crystal size 0.52×0.48×0.44 mm3.
A total of 24258 reflections, 9342 unique (Rint 0.0198)
was collected 2°BqB28°. Final R1 0.0258 (8744 data
with I\2|(I)), 0.0306 (all data), wR2 0.0603, goodness-
of-fir 1.016, final De +0.288/−0.430, absolute struc-
ture parameter 0.156(9).
3. Results and discussion
3.1. Reactions
Instead of paralleling the reactivity observed by Ker-
ber and Pakkanen for (Me2HSi)2, when (Ph2HSi)2 re-
acted with Co2(CO)8 the only isolated compound
containing a Co–Si bond was Co(SiPh3)(CO)4. This
was characterised fully by an X-ray crystal structure
What is remarkable with this reaction, in contrast to
previous examples, is the fact that it occurs sponta-

186
J.S. McIndoe, B.K. Nicholson / Journal of Organometallic Chemistry 577 (1999) 181–188
Scheme 1.
neously at room temperature without the need for
photochemical initiation, presumably because of the
higher lability of CO ligands of the Co(CO)4 group.
This means that creation of a vacant coordination site,
crucial for the 1,2-silyl migration to occur, is more
readily accomplished for the cobalt system.
Further reactions showed the generality of the trans-
formation. The mono-hydro disilanes Ph3SiSiR2H
(R2=Ph2, PhMe, Me2, Et2), Me3SiSiPh2H and PhMe2-
SiSiPhMeH were also reacted with Co2(CO)8. All ap-
peared to form mono-silyl complexes Co(SiRn-
R3% −n)(CO)4 though these were generally not separated
from each other, or from the products arising from the
eliminated silylene fragment. Evidence for mixtures was
apparent in the w(CO) IR spectra which showed
marked broadening of the peaks of the characteristic
pattern as well as small shifts in position. 1H- and
13C-NMR spectra also revealed the presence of a range
of products, typically showing three to eight signals in
the SiMe region.
came from the reaction of the germylsilanes
Ph3GeSiPh2H and Ph3GeSiEt2H with Co2(CO)8. In
both cases the product mixture was bright yellow in
colour and the w(CO) IR evidence showed characteristic
peaks assignable to Co(SiR3)(CO)4 species. The pres-
ence of some Co(GeR3)(CO)4 cannot be excluded since
it would give a similar pattern, but most of the germa-
nium was associated with other peaks which were as-
signed to the yellow germylene complex Ph2Ge-
[Co(CO)4]2, which was isolated and structurally charac-
terised. Scheme 2 suggests a possible reaction route to
these products for the example involving Ph3GeSiEt2H.
Initial formation of Co(SiEt2GePh3)(CO)4 would occur,
and a 1,2-germyl shift would give a germyl/silylene
intermediate which equilibrates with a silyl/germylene
version. The latter is likely to be the preferred form
based on the greater stability of Ge(II) compared with
Si(II). Trapping of the eliminated germylene fragment
by insertion into the Co–Co bond of Co2(CO)8 (a
known reaction for GeI2 for example) would give
Ph2Ge[Co(CO)4]2. The exclusive elimination of :GePh2,
leaving PhEt2SiCo(CO)4 is expected, since :GeR2 spe-
cies are considerably more stable than the analogous
:SiR2 species, and the Ph groups stabilise the germylene
more than Et ones. The same selectivity was found by
Pannell and Sharma [27] in the Fe(p5-C5H5)(CO)2 sys-
tem when germylsilanes were used.
In an attempt to monitor the course of the reaction,
and detect possible intermediates the reaction of
(Ph2HSi)2 with Co2(CO)8 was carried out in C6D6 in a
sealed NMR tube. Unfortunately, attempts to follow
reactions in their early stages (0–1 h) were thwarted by
extreme broadening of the peaks. This observation
suggests the presence of paramagnetic compounds, con-
sistent with the proposal of Marko´ and co-workers [26]
that the reaction between hydrosilanes and Co2(CO)8
3.2. Structural determinations
’
’
implicates Co(CO)4 and/or Co(CO)3 radicals rather
than CoH(CO)4 as proposed originally.
The crystal structure of Co(SiPh3)(CO)4 was deter-
mined as part of the definite characterisation, and
because there were no previous determinations of a
simple tri-alkyl or -aryl complex of this type, although
Co(SiH3)(CO)4, Co(SiF3)(CO)4 and Co(SiCl3)(CO)4
have been described [28–30]. The compound is iso-
morphous with Fe(PPh3)(CO)4 [31]. The structure is
shown in Fig. 1. It conforms closely to C3 symmetry,
Detection of the eliminated silylene fragment has
been achieved with other systems by trapping it with
either HSi(SiMe3)3 or with alkynes; unfortunately for
the cobalt carbonyl reactions the use of these reagents
is precluded since they both react with Co2(CO)8 them-
selves under the same conditions. However evidence for
the validity of the mechanism outlined in Scheme 1

J.S. McIndoe, B.K. Nicholson / Journal of Organometallic Chemistry 577 (1999) 181–188
187
Scheme 2.
[4] H. Tobita, K. Ueno, H. Ogino, Chem. Lett. (1990) 369.
[5] K.H. Pannell, J.M. Rozell, C. Hernandez, J. Am. Chem. Soc.
111 (1989) 4482.
[6] A. Haynes, M.W. George, M.T. Haward, M. Poliakoff, J.J.
Turner, N.M. Boag, M. Green, J. Am. Chem. Soc. 113 (1991)
2011.
[7] (a) H. Tobita, K. Ueno, H. Ogino, J. Am. Chem. Soc. 112 (1990)
3415. (b) K. Ueno, H. Tobita, H. Ogino, J. Organomet. Chem.
430 (1992) 93. (c) K. Ueno, H. Ogino, Bull. Chem. Soc. Jpn. 68
(1995) 1955.
[8] K. Ueno, K. Nakano, H. Ogino, Chem. Lett. (1996) 459.
[9] K.H. Pannell, M.C. Brun, H. Sharma, K. Jones, S. Sharma,
Organometallics 13 (1994) 1075.
[10] K.H. Pannell, H.K. Sharma, R.N. Kapoor, F. Cervantes-Lee, J.
Am. Chem. Soc. 119 (1997) 9315.
[11] W. Weidenbruch, A. Stilter, W. Saak, K. Peters, H.G. von
Schnering, J. Organomet. Chem. 560 (1998) 125.
[12] Z. Zhang, R. Sanchez, K.H. Pannell, Organometallics 14 (1995)
2605.
[13] D.C. Pestana, T.S. Koloski, D.H. Berry, Organometallics 13
(1994) 4173.
although this is not required crystallographically. The
˚
Si–Co bond length, 2.3810(7) A is equal to the longest
˚
known, 2.381(4) A for Co(SiH3)(CO)4. The equatorial
CO ligands are tilted towards the silyl substituent by 6°,
the well-known umbrella effect which is understood
theoretically [32]
The structure of the known [33] complex
Ph2Ge[Co(CO)4]2 is shown in Fig. 2, and is the first for
a simple R2GeM2 species. There are two independent
molecules in the asymmetric unit, but these do not
differ in any meaningful way. The average Ge–Co
˚
bond length of 2.47 A is longer than in Co(GeCl3)(CO)4
˚
(2.310 A) [34], but similar to those in Co(Ge-
˚
MeNpPh)(CO)4 (Np=1-naphthyl) (2.458 A) [35] and
3,4-dimethyl -1,1-bis(tetracarbonylcobaltio)-1-germa-
˚
cyclopent-3-ene (2.460 A) [36] Steric crowding at the Ge
centre in Ph2Ge[Co(CO)4]2 is indicated by the opened
Co–Ge–Co angle of 116°.
[14] (a) M.J. Michalczyk, C.A. Recatto, J.C. Calabrese, M.J. Fink, J.
Am. Chem. Soc. 114, (1992) 7955. (b) Y. Tanaka, H. Yamashita,
M. Tanaka, Organometallics 14 (1995) 530.
[15] G.P. Mitchell, T.D. Tilley, Organometallics, 15 (1996) 3477. (b)
G.P. Mitchell, T.D. Tilley, G.P.A. Yap, A.L. Rheingold,
Organometallics 14 (1995) 5472.
Acknowledgements
[16] S. Nlate, E. Herdtweck, R.A. Fischer, Angew. Chem. Int. Ed.
Eng. 35 (1996) 1861.
The University of Waikato is acknowledged for
financial support and a postgraduate scholarship (to
J.S.M.). We thank Associate Professor Cliff Rickard
and Allen Oliver, University of Auckland, for X-ray
data sets.
[17] (a) R.C. Kerber, T. Pakkanen, Inorg. Chim. Acta 37 (1979) 61.
(b) R.C. Kerber, T. Pakkanen, Inorg. Chim. Acta 49 (1981) 47.
[18] Y.L. Baay, A.G. MacDiarmid, Inorg. Chem. 8 (1969) 986.
[19] J. Greene, M.D. Curtis, Inorg. Chem. 17 (1978) 2324.
[20] G. Maas, T. Werle, in: N. Auner, U. Klingebiel (Eds.), Synthetic
Methods of Organometallic and Inorganic Chemistry, Thieme,
New York, 1996, p. 169.
[21] H.J.S. Winkler, H. Gilman, J. Org. Chem. 26 (1961) 1265.
[22] (a) W. Steudel, H. Gilman, J. Am. Chem. Soc. 82 (1960) 6129.
(b) P. Boudjouk, B.H. Han, Tetrahedron Lett. 22 (1983) 2813.
[23] R.H. Blessing, Acta Crystallogr. A51 (1995) 33.
[24] G.M. Sheldrick, SHELX97—programs for X-ray crystallography,
University of Gottingen, 1997.
References
[1] K.H. Pannell, J.R. Rice, J. Organomet. Chem. 78 (1974) 35.
[2] (a) H.K. Sharma, K.H. Pannell, Chem. Rev. 95 (1995) 1351 and
references therein. (b) M. Suginome, Y. Ito, J. Chem. Soc.
Dalton Trans. (1998) 1925.
[3] (a) K.H. Pannell, J. Cervantes, C. Hernandez, J. Cassias, S.
Vincenti, Organometallics 5 (1986) 1056. (b) H. Tobita, K.
Ueno, H. Ogino, Chem. Lett. (1986) 1777.
[25] (a) A.J. Chalk, J.F. Harrod, J. Am. Chem. Soc. 89 (1967) 1640.
(b) F.R. Anderson, M.S. Wrighton, J. Am. Chem. Soc. 106
(1984) 995.

188
J.S. McIndoe, B.K. Nicholson / Journal of Organometallic Chemistry 577 (1999) 181–188
[26] A. Sisak, F. Ungva´ry, L. Marko´, Organometallics 5 (1986) 1019.
[27] K.H. Pannell, S. Sharma, Organometallics 10 (1991) 1655.
[28] A.G. Robiette, G.M. Sheldrick, R.N.F. Simpson, B.J. Aylett,
J.A. Campbell, J. Organomet. Chem. 14 (1968) 279.
Crabtree, Organometallics 10 (1991) 3062.
[33] (a) S.A. Fieldhouse, B.H. Freeland, R.J. O’Brien, J. Chem. Soc.
D (1969) 1297. (b) M. Basato, J.P. Fawcett, A. Poe, J. Chem.
Soc. Dalton Trans. (1974) 1350.
[29] K. Emerson, P.R. Ireland, W.T. Robinson, Inorg. Chem. 9
(1970) 436.
[34] C.C. van der Berg, A. Oskam, K. Olie, J. Organomet. Chem. 80
(1974) 363.
[30] W.T. Robinson, J.A. Ibers, Inorg. Chem. 6 (1967) 1208.
[31] P.E. Riley, R.E. Davis, Inorg. Chem. 19 (1980) 159.
[32] S.A. Jackson, O. Eisenstein, J.D. Martin, A.C. Albeniz, R.H.
[35] F. Dahan, Y. Jeannin, J. Organomet. Chem. 136 (1977) 251.
[36] D. Lei, M.J. Hampden-Smith, J.W. Garvey, J.C. Huffman, J.
Chem. Soc. Dalton Trans. (1991) 2449.
.