Silylations of α,β-unsaturated and aromatic carbonyl compounds with cobalt carbonyls

Article abstract of DOI:10.1016/S0022-328X(99)00228-4

[η³-η⁵-(Silyloxy-alkenyl)]-cobalt carbonyls and various silyl ethers were formed in the cobalt carbonyl mediated silylations of α,β-unsaturated and aromatic carbonyl compounds. A silyloxonium tetracarbonylcobaltate tight ion pair was suggested as a common intermediate, which provides the products through a radical pair in the cases of aromatic aldehydes and ketones.

Full text of DOI:10.1016/S0022-328X(99)00228-4

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 586 (1999) 48–53
Silylations of a,b-unsaturated and aromatic carbonyl compounds
with cobalt carbonyls
Attila Sisak *
Department of Organic Chemistry, Research Group for Petrochemistry of the Hungarian Academy of Sciences, Uni6ersity of Veszpre´m,
Egyetem Utca 10, H-8200 Veszpre´m, Hungary
Received 4 January 1999; received in revised form 31 March 1999
Dedicated to: Professor La´szlo´ Marko´ on the occasion of his 70th birthday in recognition of his important contribution in homogeneous
catalysis.
Abstract
[h3-h5-(Silyloxy–alkenyl)]-cobalt carbonyls and various silyl ethers were formed in the cobalt carbonyl mediated silylations of
a,b-unsaturated and aromatic carbonyl compounds. A silyloxonium tetracarbonylcobaltate tight ion pair was suggested as a
common intermediate, which provides the products through a radical pair in the cases of aromatic aldehydes and ketones. © 1999
Elsevier Science S.A. All rights reserved.
Keywords: Cobalt; Silylation; Aromatic carbonyl compounds; Silicon; Ketones; Aldehydes
1. Introduction
carbonyl derivatives. The reactions also resulted in
some very interesting organic products.
a-Hydroxyalkylcobalt tetracarbonyls were proposed
several times in the last three decades as intermediates
in the hydrogenation of carbon monoxide and carbonyl
compounds in the presence of cobalt carbonyls [1–3].
We reported the formation of such complexes in reac-
tions of aldehydes and HCo(CO)4 [4,5]. These species
were found to be thermally very unstable, however,
they could be stabilized by silylating with
bis(trimethylsilyl)-trifluoroacetamide, (BSTFA) at low
temperatures leading to [a-(trialkylsilyloxy–alkyl)]-
cobalt tetracarbonyls [4,5]. The complex formed from
HCo(CO)4, crotonaldehyde and BSTFA was found to
lose CO easily giving [h3-(1-silyloxy-2-butenyl)]-cobalt
tricarbonyl (1). We obtained the same species by allow-
ing Me3SiCo(CO)4 to react with crotonaldehyde under
argon exactly as described by Murai and coworkers [6].
Now we report the silylation of several a,b-unsatu-
rated and aromatic carbonyl compounds and the prepa-
ration of some new h3-, and h5-coordinated cobalt
2. Results and discussion
The silylations were carried out using three different
methods. In method A, HCo(CO)4 was treated with an
excess of carbonyl compound at low temperature and
then a stoichiometric amount of BSTFA was added. In
method B, R3SiCo(CO)4 was prepared in situ from
Co2(CO)8 and the hydrosilane and a carbonyl com-
pound was added at 0–25°C. Catalytic quantities of
pyridine accelerated the reactions [7]. Method C was
used to synthesize Co complexes of tetraphenylcy-
clopentadienone. The result of the experiments are sum-
marized in Table 1, which along with Scheme 1 also
gives the structures of the numbered complexes.
If an organometallic complex were present, it was an
[h3-(1-silyloxy–alkenyl)]-cobalt
tricarbonyl
type
product, corresponding to the given a,b-unsaturated
carbonyl compound. With sterically very hindered ke-
tones (e.g. (+)-pulegone) no silylation took place.
Methyl vinyl ketone was allowed to react with
HCo(CO)4 at −40°C (method A). A sample of the
* Tel.: +36-88-422-022; fax: +36-88-427-492.
E-mail address: sis014@almos.vein.hu (A. Sisak)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00228-4

A. Sisak / Journal of Organometallic Chemistry 586 (1999) 48–53
49
Table 1
Silylation of a,b-unsaturated carbonyl compounds
Carbonyl compound
Crotonaldehyde c
Method a
Complex formed
Organic product formed
(yield, wt.%) b
A,B
n.d.
Methyl vinyl ketone
A
CH2ꢀCHC(OSiMe3)ꢀCH2 (9),
E,Z-CH3CHꢀC(OSiMe3)CH3 (37),
CH3CH2C(OSiMe3)ꢀCH2 (34)
d
2-Cyclohexen-1-one
Benzaldehyde
B
–
CH2CH2CH2CH2CH2CꢀO (11)
CH2CH2CH2CH2CHꢀCOSiEt2Me (31),
CHꢀCHCH2CH2CHꢀCOSiEt2Me (3),
condensation product (23)
B e
PhCH2OSiMe2Ph (85),
(PhCHOSiMe2Ph)2 (4),
PhC(OSiMe2Ph)ꢀCHOSiMe2Ph (4)
Cinnamaldehyde
Fluorenone
B e
E,Z-PhCH2CHꢀCHOSiMeEt2 (52),
E-PhCHꢀCHCH2OSiMeEt2 (22),
(PhCHꢀCH2CHOSiMeEt2)2 (15)
d
B
–
d
Dibenzosuberenone
B
–
C15H11OSiMeEt2 7 (100)
Tetraphenyl-
C b
n.d.
cyclopentadienone
a Methods used in complex preparation: see text.
b Based on GLC-analysis, see Section 3.
c See Ref. [5].
d Carbonyl compound containing Co complex has not been found as a characterizable product.
e The complex could not be isolated in pure form, see text.
reaction mixture was monitored in the temperature region
of −30 to +25°C by IR spectroscopy. At −30°C the
w(CꢁO) bands of HCo(CO)4 disappeared completely.
Beside bands belonging presumably to alkyl and acyl
complexes and Co2(CO)8, strong bands at 2052 and 1968
cm−1 were detected. The intensity of these two bands
decreased with rising temperature while those of
Co2(CO)8 increased, but they were detectable even at
room temperature (r.t.). For the unstable complex we
suggest the structure:

50
A. Sisak / Journal of Organometallic Chemistry 586 (1999) 48–53
Scheme 1.
and/or isomer, but an oxapropenyl structure proposed by
Orchin et al. [8] cannot be totally ruled out. Adding
BSTFA to the mixture at −30°C led to [2,3,4-h-{2-
methyl-2-(trimethyl–silyloxy)-butenyl}]-cobalt tricar-
bonyl (2) as the main product. No organic materials
corresponding to a radical reaction pathway were found
(see Ref. [9], vide infra).
however, further investigations are needed for the full
characterization of the complex.
Interestingly, silylation of fluorenone and dibenzo-
suberenone did not result in any p-complex; only
R3SiCo(CO)4, and Co2(CO)8 were found.
Benzaldehyde gave a h3-benzylic complex (3) which,
however, could not be isolated, but its structure is
inferred by comparing its IR spectrum with those of
related compounds1 (see Ref. [10]). The IR and 1H-NMR
spectra of the organometallic compound formed in situ
in the silylation of cinnamaldehyde were in accord with
an h3-allyl structure (4).
In the case of tetraphenylcyclopentadienone, a cyclic
dienone, [h5-(1-silyloxy-2,3,4,5-tetraphenylcyclopentadi-
enyl)]-cobalt dicarbonyl (5) was prepared and character-
ized. Another, less soluble species could be isolated as
well. The latter IR spectrum showed not only two
terminal, but also an organic w(CꢁO) band. It proved to
be paramagnetic by NMR and magnetic measurements,
however, no signal in its ESR spectrum could be found
in solution. On the basis of IR, MS and elementary
analytical data we propose the structure:
Most of the silylations were performed in the presence
of an excess (5- to 20-fold) of carbonyl compound, i.e.
the reactions became quasicatalytic, and organic silyla-
tion products were found. It can be seen from Table 1
that the organic products from the aliphatic (or cy-
cloaliphatic) and those from the aromatic (or vinylog)
carbonyl compounds were different. Starting from the
members of the first group the corresponding silyl enol
ethers and silyl dienol ethers were formed. (see Table 1).
2-Cyclohexen-1-one also gave some condensation prod-
ucts.
On the other hand, the silylation of aromatic carbonyl
compounds led to, among others, disilyl pinacol ethers.
The highest selectivity for this compound was generated
from fluorenone, where {[9,9%-bi-9H-fluorene]-9,9%-diyl-
bis(oxy)}-bis(diethylmethyl-silane) (6) (see Ref. [11]), was
the only characterizable product. In contrast, dibenzo-
suberenone provided only monosilyl ether 7 as product.
Both types of compounds were found in the case of
cinnamaldehyde, but from benzaldehyde several byprod-
ucts were formed. One of them, 1,2-bis(diethyl-
methyl–silyloxy)-2-phenyl-ethene might come from
1 The IR spectrum could be assigned by analogy to known h3-ben-
zylcobalt complexes, which were prepared according to the equation:

A. Sisak / Journal of Organometallic Chemistry 586 (1999) 48–53
51
3. Experimental
carbonylation side reactions because its analog was
found in the hydrosilylation of benzaldehyde under
CO by Murai and coworkers [12].
3.1. General
In our earlier study on the cobalt carbonyl medi-
ated silylation of aldehydes [5] and oxiranes [13] we
suggested that hydrido- or silylcobalt tetracarbonyl
form an ionic intermediate with the substrate.
Analogous suggestions were made by Gladysz and
coworkers for the reaction of Me3SiMn(CO)5 with
aliphatic and aromatic aldehydes and ketones [14].
These authors found a similarly different nature of
the organic silylation products of aliphatic and aro-
matic compounds. From benzaldehyde, e.g. unstable
Me3SiOCH(C6H5)Mn(CO)5 was formed which decom-
posed to the corresponding bis-(trimethylsilyl) pinacol
ether. Aliphatic aldehydes or ketones gave, however,
only HMn(CO)5 and silyl enol ethers. It has been
shown that both types of organic products could be
produced catalytically by cobalt carbonyls even if the
selectivity of silyl pinacol ethers varied strongly with
the structure of aromatic carbonyl compounds (see
Refs. [5,12] and this work).
Based on our former and current results as well as
literature analogies, we propose a mechanism of the
cobalt carbonyl mediated silylation of a,b-unsaturated
carbonyl compounds as shown in Scheme 1. The cru-
cial point of the mechanism is the transformation of
the, presumably tight, silyloxonium tetracarbonyl-
cobaltate ion pair 8 either to an [h1-(1-silyloxy–
alkenyl)]-cobalt tetracarbonyl intermediate (9), or to a
radical pair 10 (see Ref. [12]). At low CO concentra-
tions, 9 may lose CO and an h3- (or h5-) silyloxy–
alkenyl complex will be formed. The collapse of 10
results again in the formation of 9, the escape, how-
ever, provides solvent separated radicals. Co(CO)4
will recombine to give Co2(CO)8 making a catalytic
cycle possible. The substituted silyloxymethyl radical
can react with the hydrosilane to a saturated silyl
ether, or recombine to disilyl pinacol ether.
HCo(CO)4 formed by b-elimination decomposes un-
der the reaction conditions (see Ref. [15]).
All manipulations involving air-sensitive compounds
were carried out by the usual Schlenk technique using
deoxygenated, dry solvents and gases, and reaction
vessels with magnetic stirring. IR spectra were
recorded by using a 0.06 mm CaF2 cuvette on a
Specord IR 75 (Carl Zeiss, Jena) spectrometer, which
was calibrated with benzene (1959.6 cm−1
) and
1
polystyrene (1601.4 cm−1). H- and 13C-NMR spectra
were obtained on a Varian Unity 300 spectrometer,
using hexamethyldisiloxane internal reference. Gas
chromatograms were recorded on a Hewlett–Packard
model 5830 A chromatograph, GLC-MS analyses
were performed on a JMS 01-SG-2 Jeol spectrometer
and a Hewlett–Packard 5890 Series II GC-MSD
equipment.
Solutions of HCo(CO)4 in pentane or octane were
prepared from Co2(CO)8, DMF, and concentrated
HCl [16], and contained 1–3 mol% Co2(CO)8, accord-
ing to IR and Co analyses. Me3SiH [17] and
R3SiCo(CO)4 complexes (R3=Me3, Et2Me, Et3,
Me2Ph) [18] were prepared by literature methods.
Other starting materials were commercial products
and were purified by crystallization or distillation.
Most of the organic products (see Table 1) are
known silyl ethers or their analogs (e.g. diethyl-
methyl–silyl derivatives instead of trimethylsilyl ones).
They were identified on the basis of known MS spec-
tra. In the absence of such data authentic samples
were
prepared
by
described
methods
[11,12,14,19,20,22] and the GLC-MS data were com-
pared wit those of the original samples.
3.2. Silylation procedures
3.2.1. Method A
To a cold (−40°C) solution of 190 ml (2.28 mmol)
of methyl vinyl ketone in 5 ml of dichloromethane
1.5 ml of a 0.7 M stock solution of HCo(CO)4 (1.26
mmol in octane) was added under carbon monoxide.
The reaction was followed by IR spectroscopy (see
text). The solution was allowed to warm up to −
30°C and 175 ml of BSTFA (0.66 mmol) was added.
After freezing out Co2(CO)8, [2,3,4-h-{2-methyl-2-
(trimethylsilyloxy)-butenyl}]-cobalt tricarbonyl (2) as
the main product was detected by IR spectroscopy
(w(CꢁO): 2041 (vs), 1969 (vs), see Ref. [6]).
MS-data (m/z) of the identified organic products (see
Refs. [19,20]).
Aliphatic and cycloaliphatic a,b-unsaturated car-
bonyl compounds can be transformed to [h3-(1-silyl-
oxy–alkenyl)]-cobalt tricarbonyl derivatives (1, 2)
through a stoichiometric reaction with HCo(CO)4 and
BSTFA (see Ref. [5]). as was demonstrated in the
case of methyl vinyl ketone (vide supra). The proba-
bility of the two ways leading to the product (Scheme
1) may depend on CO concentration.
We expected that in the case of aromatic aldehydes
and ketones a CIDNP effect could be detected fol-
1
2-Trimethylsilyloxy-1,3-butadiene: 142 (M+), 127,
85, 75, 73, 45.
lowing their silylation by H-NMR spectroscopy (see
Ref. [14]). In the course of in situ silylation of ben-
zaldehyde and cinnamaldehyde in an NMR tube
however, no CIDNP effect has been observed.
2-Trimethylsilyloxy-2-butene, (E,Z): 144 (M+), 129,
75, 73, 45.

52
A. Sisak / Journal of Organometallic Chemistry 586 (1999) 48–53
2-Trimethylsilyloxy-1-butene: 144 (M+), 129, 75, 73,
recorded at 25°C. No really well-resolved spectra were
obtained, presumably due to transient radical species (see
text). No CIDNP effect could be observed, the following
new signals were found (l, ppm (rel. int.)): 1.82 (B1),
2.66 (:1), 3.25 (:1), 3.45 (:2), 4.35 (:1), 4.66 (:1),
5.20 (B1), 5.57 (:1), 6.35 (:2), 6.62 (B1), 7.3 (broad,
\10).
45.
3.2.2. Method B
Up to 1.0 mmol of Co2(CO)8 was dissolved in 5–10
ml of dichloromethane or toluene (in the case of benz-
aldehyde—pentane) under carbon monoxide. Carbonyl
compound (10–30 mmol) and 7–20 mmol of hydrosilane
and—occasionally—up to 1 mmol of base (pyridine or
triethyl amine) was added at 0°C. The reaction mixture
was stirred for 0.5–2 h at 0–25°C (before the end of
stirring at least 10 min at r.t.). Probe(s) for IR analysis
were taken from the reaction mixture. At the end of the
stirringCo2(CO)8 andR3SiCo(CO)4 wereremovedadding
excess of pyridine and the organic products were analyzed
by GLC-MS. Partial results are shown in Table 1, others
are listed here:
Six of these signals corresponds to the reported spec-
trum
of
1-(trimethyl–silyloxy)-3-phenyl-1-propene
(E:Z:2:1) [22]. The signals at 1.82, 2.66 and 5.57 ppm
may be assigned to 4 (see Refs. [6,10]) and those at 4.35,
6.35 (partly) and 6.62 to the third silyl ether isomer,
respectively. The broad phenyl signal at 7.3 ppm belongs
to all four components. GLC analysis of the reaction
mixture showed an actual isomer ratio of [E
-
PhCH2CH ꢀ CHOSiEt3]:[Z-PhCH2CH ꢀ CHOSiEt3]:
[PhCHꢀCH–CH2OSiEt3]:8:4:7. Disilyl pinacol ether
was found only in traces under the above conditions.
MS data (m/z) of the organic products of the catalytic
process (see Ref. [22]): 1-(diethylmethyl–silyloxy)-3-
phenyl-2-propene, (E,Z): 234 (M+), 205, 149, 121, 117,
91, 73. 1-(Diethylmethyl–silyloxy)-3-phenyl-1-propene,
(E): 234 (M+), 205, 149, 117, 115, 91, 61. 1,2-
Bis(diethylmethyl – silyloxy)-1,2-bis(1-phenyl-ethenyl)-
ethane (tentative assignment): 233, 175, 101, 73, 45.
3.2.2.1. 2-Cyclohexene-1-one. IR: [Py+SiEt2Me.
Co(CO)4−] tight ion pair [21] could be detected. MS data
(m/z) of the organic products (see Refs. [9,20]). (Diethyl-
methyl–silyloxy)-cyclohexene, (E,Z): 198 (M+), 169,
156, 142, 89, 61. (Diethylmethyl–silyloxy)-1,3-cyclohexa-
diene: 196 (M+), 165, 137, 73, 61.
Another species with M+ =294 was found, which may
be an aldol-condensations product.
3.2.2.4. Fluorenone. IR: only Et2MeSiCo(CO)4 and
Co2(CO)8 couldbedetectedascobaltcontainingproducts.
MS of {[9,9%-bi-9H-fluorene]-9,9%-diylbis(oxy)}-bis(di-
ethylmethyl–silane) (6) (see Ref. [12]), (m/z): 281, 251,
223, 165, 101, 73, 45.
3.2.2.2. Benzaldehyde. IR: an h3-type complex was found
as the only organometallic species, but it could not be
isolated in pure form: w(CꢁO) (pentane): 2048 (vs), 1988
(vs), 1967 (vs). We assign the spectrum tentatively to
[h3-{phenyl-(dimethylphenyl–silyloxy)-methyl}]-cobalt
tricarbonyl (3).
MS data (m/z) of the organic products (see Refs.
[12,14]). (Dimethylphenyl–silyl) benzyl ether: 242 (M+),
227, 164, 149, 91. 1,2-Bis(dimethylphenyl–silyloxy)-1,2-
diphenyl-ethane, (erythro–threo): 271, 241, 167, 135, 91.
1,2-Bis(dimethylphenyl–silyloxy)-1-phenyl-ethene: 404
(M+), 330, 209, 179, 135.
3.2.2.5. Dibenzosuberenone. IR: see Section 3.2.2.4.
MS of 5-(diethylmethyl–silyloxy)-5H-dibenzo[a,d]-
cycloheptene (7) (m/z): 308 (M+), 274, 206, 190, 188, 161,
105, 89, 73.
3.2.3. Method C
From the reaction mixture of 0.58 g (1.54 mmol) of
Co2(CO)8 and 0.5 ml (3.45 mmol) of Et2MeSiH the excess
of silane was removed in vacuo. To the resulting
Et2MeSiCo(CO)4, 20 ml of toluene, 1.3 g (2.60 mmol) of
tetraphenyl–cyclopentadienone and 80 ml (1.0 mmol) of
pyridine were added. The reaction mixture was warmed
to 50°C. The gas evolution ceased within 1.5 h. The
brown–violet mixture was then cooled successively to
−78°C. A brown–violet species could be frozen out
which needs further characterization. The wine-red
mother liquor was evaporated to dryness in vacuo. The
residue was recrystallized twice from hexane which
resulted in bright-red microcrystals which proved to be 5.
3.2.2.3. Cinnamaldehyde. IR: the expected h3-allyl-type
complex was found in the reaction mixture but it could
not be isolated: w(CꢁO) (toluene): 2047 (vs), 1984 (vs),
1973 (vs). We assigned the spectrum tentatively to [1,2,3
-h- {1 - phenyl - 3 - (diethylmethyl – silyloxy)propenyl}]-
cobalt tricarbonyl (4) (more isomers are possible).
1H-NMR: the following procedure was used to test for
CIDNP effect as well as to characterize the in situ formed
products.
A total of 17 mg (0.05 mmol) of Co2(CO)8 and 38 ml
(0.25 mmol) of Et3SiH were reacted at 0°C under Ar in
a Schlenk-tube. As H2 evolution ceased, the reaction
mixture was transferred in an NMR tube with a syringe,
the tube was chilled at −80°C. Then, 0.45 ml of CD2Cl2
and 16 ml (0.13 mmol) of freshly distilled trans-cin-
namaldehyde were added. 1H-NMR spectra were
3.2.3.1. [p5-{(1-Diethylmethyl–silyloxy)-(2,3,4,5-tetra-
phenyl)-cyclopentadienyl}]-cobalt dicarbonyl (5). IR:
w(CꢁO) (hexane): 2013 (vs), 1953 (vs).
MS (m/z): 600 (M+), 572 (M+ −CO), 544 (M+
−2CO), 515 ([C5(C6H5)4Co]O+Si(C2H5)2CH3), 488

A. Sisak / Journal of Organometallic Chemistry 586 (1999) 48–53
53
Dombek, Adv. Catal. 32 (1983) 325.
[3] J. Falbe, New Syntheses with Carbon Monoxide, Springer,
New York, 1980.
([C5(C6H5)4Co]OSi(CH3)), or (C5(C6H5)4HOSi(C2H5)2-
CH3), 101 (Si+(C2H5)2CH3), 73 (base peak, HSi+
(C2H5)CH3), 59 (Co).
[4] A. Sisak, E. Szerencse´s-Sa´mpa´r, V. Galamb, L. Ne´meth, F.
Ungva´ry, G. Pa´lyi, Organometallics 8 (1989) 1096.
[5] F. Ungva´ry, A. Sisak, L. Marko´, Adv. Chem. Ser. 230 (1992)
297.
[6] Previously, Murai and coworkers published the formation of
[h3-(1-silyloxy-propenyl)]cobalt tricarbonyl using acrolein under
similar conditions. N. Chatani, Y. Yamasaki, S. Murai and N.
Sonoda, Tetrahedron Lett. 24 (1983) 5649.
1H-NMR (l, ppm, CDCl3): 0.07 (s, SiCH3
6
, 3H), 0.23
(q, CH2
6
, J 4.6 Hz, 4H), 0.68 (t, CH3
, 20H).
6 , J 4.6 Hz, 6H),
6.95–7.46 (m, C6H5
6
13C-NMR (l, ppm, CDCl3): −4.26, 1.01, 6.29,
96.75, 98.53, 127.14 (2C: 6, 7), 127.38 (3C: 8, 9),
127.70, 131.60, 131.98, 132.21, 132.48, 205.55.
[7] If the reaction were performed in the presence of amine
(cobalt:amine=1:1) it started at 0°C (see Section 3).
[8] R.W. Goetz, M. Orchin, J. Am Chem. Soc. 85 (1963) 2782.
[9] J. Roth, K. Grega, M. Orchin, J. Organomet. Chem. 342
(1988) 129.
[10] (a) V. Galamb, G. Pa´lyi, F. Ungva´ry, L. Marko´, R. Boese, G.
Schmid, J. Am. Chem. Soc. 108 (1986) 3344. (b) C. Zucchi, G.
Pa´lyi, V. Galamb, E. Sa´mpa´r- Szerencse´s, L. Marko´, P. Li, H.
Alper, Organometallics 15 (1996) 3222.
[11] J.V. Crivello, D.A. Conlon, J.L. Lee, J. Polym. Sci. Part A 24
(1986) 1197.
Compound 5 gave correct elemental analysis.
[12] N. Chatani, H. Tokuhisa, I. Kokubu, S. Fujii, S. Murai, J.
Organomet. Chem. 499 (1995) 193.
[13] J. Kreisz, A. Sisak, L. Marko´, F. Ungva´ry, J. Organomet.
Chem. 451 (1993) 53.
Acknowledgements
The authors acknowledge helpful discussions with
Professor F. Ungva´ry and Drs Ga´bor Szalontai, Zolta´n
De´csi, and Bernadett Sz. Ravasz, as well as financial
support from the Hungarian Academy of Sciences
(Grant no. OTKA T4211/1992).
[14] D.L. Johnson, J.A. Gladysz, Inorg. Chem. 20 (1981) 2508.
[15] F. Ungva´ry, L. Marko´, J. Organomet. Chem. 193 (1980) 383.
[16] I. Wender, H.W. Sternberg, M. Orchin, J. Am. Chem. Soc. 75
(1953) 3041.
[17] H. Sakurai, K. Miyoshi, Y. Nakadaira, Tetrahedron Lett.
(1977), 2671.
[18] A.J. Chalk, J.F. Harrod, J. Am. Chem. Soc. 89 (1967) 1640.
[19] L. Hansson, R. Carlsson, Acta Chim. Scand. 43 (1989) 188.
[20] H.O. House, L.J. Czuba, M. Gall, H.D. Olmstead, J. Org.
Chem. 34 (1969) 2324.
References
[21] A. Sisak, F. Ungva´ry, L. Marko´, Organometallics 5 (1986)
1019.
[22] J.K. Rasmussen, A. Hassner, J. Org. Chem. 39 (1974) 2558.
[1] L. Aldridge, B. Jonassen, J. Am. Chem. Soc. 85 (1963) 886.
[2] (a) W. Keim (Ed.) in: Catalysis in C1 Chemistry, Reidel, Dor-
drecht, 1983, pp. 5–39. and references therein. (b) B.D.
.