Formation and structure of heptacoordinate dihalogeno carbene-C,O chelate complexes and dihalogeno carbene-C,O chelate phosphine complexes of molybdenum(II) and tungsten(II): A new class of fischer type carbene complexes

Article abstract of DOI:10.1021/om010429e

The carbene-C,O chelate carbonyl tungsten(O) complex [(CO)₄W=C(OMe)C₆H₄OMe-o] (1) reacts with phosphines by opening of the chelate ring to form [(CO)₄(PR₃)W=C(OMe)C₆H₄-OMe-o] (R = Me (5), ⁿBu (6), Ph (7), Cy (8)). Treatment of 5 and 6 with SnX₄ affords by oxidative decarbonylation, depending on R and X, either one or two isomers of the carbene-C,O chelate dicarbonyl phosphine tungsten(II) complexes [X₂(CO)₂(PR₃)W=C(OMe)C₆H₄ OMe-o] (R = Me, X = Cl (9a), Br (9b), I (9c-A/9c-B); R = ⁿBu, X = Cl (10a), Br (10b-A/10b-B), I (10c)). The isomers do not interconvert. In the corresponding reaction of 7 with SnBr₄ instead of a carbene-C,O chelate dicarbonyl phosphine tungsten(II) complex the carbene-C,O chelate tricarbonyl tungsten(II) complex [Br₂(CO)₃W=C(OMe)C₆H₄OMe-o] (3b) is formed. When 8 is treated with SnBr₄, the formation of a carbene-C,O chelate phosphine tungsten-(II) complex cannot be detected any more. Like the o-OMe tungsten(O) complexes 5 and 6 the p-OMe tungsten(O) complex [(CO)₄(PⁿBu₃)W=C(OMe)C₆H₄ OMe-p] (11) and the ring-unsubstitut complex [(CO)₄(PⁿBu₃)W=C(OMe)Ph] (13) react with SnBr₄ by oxidative decarbonylation and formation of W(II) complexes, [Br₂(CO)₃(PⁿBu₃)W=C(OMe)R′] (R′ = C₆H₄OMe-p, Ph) (two isomers each). Dicarbonyl carbene-C,O chelate trimethylphosphine metal(II) complexes [X₂(CO)₂(PMe₃)M=C(OMe)C₆ H₄OMe-o] (M = Mo, W; X = Cl, Br) are also accessible by substitution of PMe₃ for a CO ligand in [X₂(CO)₃M=C(OMe)C₆H₄OMe-o]. However, when PMe₃ is replaced by the bulkier PPh₃ in the reaction with [Br₂(CO)₃-M=C(OMe)C₆H₄OMe-o], instead of substitution, the opening of the chelate ring is observed.

Full text of DOI:10.1021/om010429e

4040
Organometallics 2001, 20, 4040-4048
F or m a tion a n d Str u ctu r e of Hep ta coor d in a te Dih a logen o
Ca r ben e-C,O Ch ela te Com p lexes a n d Dih a logen o
Ca r ben e-C,O Ch ela te P h osp h in e Com p lexes of
Molybd en u m (II) a n d Tu n gsten (II): A New Cla ss of
F isch er Typ e Ca r ben e Com p lexes
Ru¨diger Stumpf, Monika J aeger, and Helmut Fischer*
Fachbereich Chemie, Universta¨t Konstanz, Fach M727, 78457 Konstanz, Germany
Received May 22, 2001
The carbene-C,O chelate carbonyl tungsten(0) complex [(CO)4WdC(OMe)C6H4OMe-o] (1)
reacts with phosphines by opening of the chelate ring to form [(CO)4(PR3)WdC(OMe)C6H4-
n
OMe-o] (R ) Me (5), Bu (6), Ph (7), Cy (8)). Treatment of 5 and 6 with SnX4 affords by
oxidative decarbonylation, depending on R and X, either one or two isomers of the carbene-
C,O chelate dicarbonyl phosphine tungsten(II) complexes [X2(CO)2(PR3)WdC(OMe)C6H4OMe-
n
o] (R ) Me, X ) Cl (9a ), Br (9b), I (9c-A/9c-B); R ) Bu, X ) Cl (10a ), Br (10b-A/10b-B),
I (10c)). The isomers do not interconvert. In the corresponding reaction of 7 with SnBr4
instead of a carbene-C,O chelate dicarbonyl phosphine tungsten(II) complex the carbene-
C,O chelate tricarbonyl tungsten(II) complex [Br2(CO)3WdC(OMe)C6H4OMe-o] (3b) is formed.
When 8 is treated with SnBr4, the formation of a carbene-C,O chelate phosphine tungsten-
(II) complex cannot be detected any more. Like the o-OMe tungsten(0) complexes 5 and 6
the p-OMe tungsten(0) complex [(CO)4(PnBu3)WdC(OMe)C6H4OMe-p] (11) and the ring-
unsubstituted complex [(CO)4(PnBu3)WdC(OMe)Ph] (13) react with SnBr4 by oxidative
decarbonylation and formation of W(II) complexes, [Br2(CO)3(PnBu3)WdC(OMe)R′] (R′ )
C6H4OMe-p, Ph) (two isomers each). Dicarbonyl carbene-C,O chelate trimethylphosphine
metal(II) complexes [X2(CO)2(PMe3)MdC(OMe)C6H4OMe-o] (M ) Mo, W; X ) Cl, Br) are
also accessible by substitution of PMe3 for a CO ligand in [X2(CO)3MdC(OMe)C6H4OMe-o].
However, when PMe3 is replaced by the bulkier PPh3 in the reaction with [Br2(CO)3-
MdC(OMe)C6H4OMe-o], instead of substitution, the opening of the chelate ring is observed.
In tr od u ction
Usually, the oxidation of carbene carbonyl complexes
leads to a cleavage of the metal-carbene bond and the
former carbene ligand (a) dimerizes to give an olefin,3
(b) is transformed into the corresponding carbonyl
compound,4 or (c) provides imidazolidin-2-ylidium salts
(as has been observed in the reactions of some cyclic
diamino-substituted group VI metal carbene complexes
with iodine).5 We recently observed that under certain
conditions Fischer carbene complexes can be oxidized
without cleavage of the metal-carbene bond, giving rise
to the formation of rather rare metal(II) carbene com-
plexes. Thus, the carbene-C,O chelate carbonyl com-
plexes 1 and 2 are readily oxidized by tin(IV) halides
or halogens to give the carbene-C,O chelate carbonyl
The oxidative decarbonylation of molybdenum or
tungsten carbonyl complexes is a very common and well-
known reaction. Typically, the hexacoordinate metal-
(0) center is oxidized by two units and one carbonyl
ligand is replaced by two halides, resulting in heptaco-
ordinate metal centers. Extensive studies in this field
have been performed by Baker et al., and two reviews
in this field have been published recently.1
Halogeno carbonyl complexes with additional carbene
ligands are thought to play an important role as
intermediates in the metathesis of olefins with halogeno
carbonyl complexes as precursors.2
* To whom correspondence should be addressed. Tel.: +7531-
882783. Fax: +7531-883136. E-mail: helmut.fischer@uni-konstanz.de.
(1) For recent reviews see: (a) Baker, P. K. Adv. Organomet. Chem.
1996, 40, 45-115. (b) Baker, P. K. Chem. Soc. Rev. 1998, 27, 125-
131.
(2) (a) Bencze, L.; Marko´, L. J . Organomet. Chem. 1971, 28, 271-
272. (b) Bencze, L.; Marko´, L. J . Organomet. Chem. 1974, 69, C19-
C20. (c) Bencze, L.; Kraut-Vass, A.; Pro´kai, L. J . Chem. Soc., Chem.
Commun. 1985, 911-912. (d) Bencze, L.; Kraut-Vass, A. J . Mol. Catal.
1985, 28, 369-380.
(3) See, e.g.: (a) Le Bozek, H.; Dixneuf, P. H. J . Chem. Soc., Chem.
Commun. 1983, 1462. (b) Moinet, C.; Le Bozek, H.; Dixneuf, P. H.
Organometallics 1989, 8, 1493.
(4) See, e.g.: (a) So¨derberg, B. C.; Bowden, B. A. Organometallics
1992, 11, 2220. (b) Licandro, E.; Maiorana, S.; Papagni, A.; Zanotti
Gerosa, A.; Cariati, F.; Bruni, S.; Moret, M.; Chiesi-Villa, A. Inorg.
Chim. Acta 1994, 220, 233.
(5) Liu, S.-T.; Ku, R.-Z.; Liu, C.-Y.; Kiang, F.-M. J . Organomet.
Chem. 1997, 543, 249.
10.1021/om010429e CCC: $20.00 © 2001 American Chemical Society
Publication on Web 08/17/2001

A New Class of Fischer Type Carbene Complexes
Organometallics, Vol. 20, No. 19, 2001 4041
dihalogeno molybdenum and tungsten complexes 3a -c
and 4a ,b, respectively (eq 1).6
The PCy3 complex 8 was obtained as a mixture of the
cis and the trans isomers (ratio ca. 4.5:1). All other
complexes were formed highly stereoselectively, with the
cis isomer being detected exclusively. In contrast, in
earlier experiments the substitution of PR3 for a CO
ligand in alkylcarbene pentacarbonyl chromium and
tungsten complexes was observed to afford mixtures of
the cis and trans isomers. It was possible to separate
the isomers by column chromatography. In solution
above 30 °C a slow isomerization to form cis/trans
mixtures again was obtained. The cis/trans equilibrium
ratio was found to depend mostly on the steric require-
ments of the carbene and the phosphine ligand.9
At low temperature, alkylphosphines add to the
carbene carbon atom of alkylcarbene complex to give
ylide complexes.10 The addition is reversible.11 In con-
trast, there was no indication for the formation of ylide
complexes when phosphines were added to solutions of
1. Presumably, the equilibrium (carbene complex +
PR3)/ylide complex is too far on the side of the carbene
complex for the ylide complexes to be detectable.
A molybdenum complex related to 7 was generated
by Do¨tz et al. by reaction of 2 with PPh3 at low
temperature. The resulting carbene tetracarbonyl phos-
phine complex was labile and decomposed already below
0 °C to regenerate 2 and PPh3.12 When at room tem-
perature PPh3 was used in excess, the substitution of
PPh3 for a CO ligand was observed. In contrast, the
complexes 5-8 are stable at room temperature.
Oxid a t ion of t h e Ca r b en e P h osp h in e Com -
p lexes. Similar to the carbene-C,O chelate complex 1
the trimethylphosphine complex 5 quickly reacted with
SnBr4 by evolution of a gas and formation of a white
precipitate. As judged by the NMR spectrum of the
solution, only one complex (9b) was formed quantita-
tively. It was not possible to purify the new compound
9b by column chromatography, since it irreversibly
adsorbed on silica. However, filtration through Celite
(to remove the white solid SnBr2) and recrystallization
from CH2Cl2/pentane afforded pure 9b in 84% yield (eq
3). Complex 9b was also formed, in addition to other
Neither the synthesis of a carbene-C,O chelate diiodo
molybdenum complex nor that of carbene-C,O chelate
dihalogeno chromium complexes could be achieved by
this route.6 When SnCl4 was added to solutions of the
carbene-C,O chelate tetracarbonyl chromium complex
related to 1 and 2, only a slow decomposition of the
carbene complex and formation of [Cr(CO)6] was ob-
served. Likewise, no reaction between SnCl4 and the
nonchelating o-anisylcarbene pentacarbonyl complex
[(CO)5WdC(OMe)C6H4OMe-o] was observed. In con-
trast, [(CO)5WdC(OMe)Ph] did react with SnCl4; how-
ever, the reaction afforded, among other products, a
carbyne complex.6
From these observations it was concluded that an
energetically high-lying HOMO at the metal is required
for the oxidative decarbonylation of carbene carbonyl
complexes by SnCl4 or halogens.
Resu lts
Syn th esis of Ca r ben e P h osp h in e Com p lexes. To
test this hypothesis and to determine whether chelation
is a prerequisite for the oxidative decarbonylation,
nonchelating carbene tetracarbonyl phosphine com-
plexes were assumed to be suitable starting complexes.
In earlier studies, several carbene tetracarbonyl phos-
phine complexes were prepared either by thermolysis
or by photolysis of carbene pentacarbonyl complexes in
the presence of phosphines.7
The carbene tetracarbonyl phosphine complexes 5-8
were obtained by addition of phosphines to solutions of
the carbene-C,O chelate carbonyl tungsten complex 1,
thus establishing that an opening of the chelate ring
can be induced by nucleophiles. The complexes were
formed almost quantitatively and, after column chro-
matography, were isolated in high yields (eq 2). Complex
7 has been prepared before.8
unidentified products, when 5 was treated with bromine
in CDCl3. However, it is more convenient to prepare 9b
by reaction of 5 with SnBr4. This method gives some-
what higher yields, and the absence of byproducts
facilitates purification of 9b.
(9) (a) Fischer, E. O.; Fischer, H.; Werner, H. Angew. Chem. 1972,
84, 682-683; Angew. Chem., Int. Ed. Engl. 1972, 11, 644-645. (b)
Fischer, H.; Fischer, E. O. Chem. Ber. 1974, 107, 673-679.
(10) Kreissl, F. R.; Fischer, E. O.; Kreiter, C. G.; Fischer, H. Chem.
Ber. 1973, 106, 1262-1276.
(11) (a) Fischer, H.; Fischer, E. O.; Kreiter, C. G.; Werner, H. Chem.
Ber. 1974, 107, 2459-2467. (b) Fischer, H. J . Organomet. Chem. 1979,
170, 309-317.
(12) Do¨tz. K. H.; Larbig, H.; Harms, K. Chem. Ber. 1992, 125, 2143-
2148.
(6) J aeger, M.; Stumpf, R.; Troll, C.; Fischer, H. Chem. Commun.
2000, 931-932.
(7) Fischer, E. O.; Fischer, H. Chem. Ber. 1974, 107, 657-672.
(8) J aeger, M. Dissertation, Universita¨t Konstanz, 1993.

4042 Organometallics, Vol. 20, No. 19, 2001
Stumpf et al.
Analogously, addition of SnCl4 to solutions of 5 led to
the formation of complex 9a (eq 3). Compound 9a was
not isolated but only identified by its IR and NMR
spectra since, on a preparative scale, it was more easily
accessible by a different route (vide infra). Surprisingly,
two carbonyl ligands are lost in the course of the
formation of 9a and 9b from 5 and SnX4.
Both new complexes were characterized by IR and
NMR spectroscopy, and complex 9b was additionally
characterized by a single-crystal X-ray analysis. How-
ever, the crystals were of only poor quality and the PMe3
ligand was severely disordered. Therefore, a detailed
discussion of the distances and angles in 9b is not
feasible. Nevertheless, the following structural features
were unambiguously established. The PMe3 ligand
assumes a position almost trans to the coordinating
OMe group. Complex 9b contains a mirror plane which
renders the two bromo and the two carbonyl ligands
equivalent, in accord with the observation of only one
13C resonance for the CO ligands.
The oxidative decarbonylation of 5 with SnI4 deviated
from those reactions involving SnCl4 and SnBr4. Instead
of one complex, two new complexes were formed. It was
not possible to separate them by crystallization, and
attempted column chromatography led only to irrevers-
ible adsorption on silica. From the NMR spectrum of
the reaction mixture and the elemental analysis of the
isolated product mixture it followed that the oxidation
of 5 with SnI4 gave the two isomeric complexes 9c-A
and 9c-B (eq 4). A rapid equilibration of the two isomers
Similar to the case for 5, the tributylphosphine
complex 6 readily reacted with SnX4 by oxidative
decarbonylation and chelation. Again, the structure of
the product complexes strongly depended on X. The
reaction of 6 with SnCl4 produced only one complex
(10a ), whose structure is similar to that of 9a , 9b, and
9c-A. In contrast to 5, treatment of 6 with SnBr4 gave
a mixture of two isomeric complexes, 10b-A and 10b-
B, and in addition small amounts of a third unidentified
complex. The mixture could not be separated. On the
basis of the IR and NMR spectra 10b-A was assigned a
structure similar to those of 9a , 9b, 9c-A, and 10a .
Isomer 10b-B in turn is structurally related to 9c-B.
The complexes 10b-A and 10b-B were formed in the
ratio 10b-A/10b-B ≈ 2:1. Neither broadening nor a
change in the relative intensities of the resonances in
1
the H NMR spectrum was observed when solutions of
the isomeric mixture were warmed to 60 °C. Therefore,
an interconversion of the isomers could be excluded.
The reaction of 6 with SnI4 again afforded only one
isomer. However, in contrast to 10a , complex 10c is
structurally related to 9c-B (Scheme 1).
When the steric requirements of the phosphine ligand
were further increased, the reactions of the carbene
tetracarbonyl phosphine complexes with tin tetrahalides
took a different course. The reaction of the triphen-
ylphosphine complex 7 with SnBr4 in CDCl3 did not
afford a carbene-C,O chelate phosphine complex but
rather the tricarbonyl complex 3b (eq 5). Thus, in the
course of the reaction, instead of two carbonyl ligands,
only one carbonyl and the phosphine ligand were
replaced. Obviously, the very bulky PPh3 (cone angle
145°) prevents the accommodation of seven ligands at
the tungsten(II) atom in a stable arrangement.
Complex 3b was identified by comparison of its
spectra with those of an authentic sample.6 Two equiva-
lents of SnBr4 was required to drive the reaction to
completion. One equivalent of SnBr4 was consumed in
the formation of the adduct SnBr4‚PPh3, whose oxida-
tion potential is insufficient to oxidize the carbene
complex 7. Adduct formation and thus reduction of the
oxidation potential of SnX4 also prevents the use of THF
as the solvent. Tin tetrahalides are known to form with
THF adducts, [SnX4(thf)2].13
Like 7, the ring-unsubstituted complex cis-[(CO)4-
(PPh3)WdC(OMe)Ph] also reacted with SnBr4. However,
a dibromo carbene tungsten(II) complex could not be
detected among the products. NMR analysis of the
reaction mixture indicated loss of the carbene ligand
from the complex.
Chelation through a nucleophile such as, for example,
OMe in 3b seemed to be an important factor in stabiliz-
ing tungsten(II) carbene complexes. Therefore, we next
could be excluded, since the ratio remained unaffected
when solutions of 9c-A and 9c-B were cooled or heated.
The complexes 9c-A and 9c-B were also obtained in
slightly different ratios (9c-A:9c-B is approximately 1:2)
when the conditions for the reaction of 5 with SnI4 were
varied. A comparison of the IR spectra of 9c-A/9c-B with
those of the isomerically pure complexes 9a and 9b
indicated that the structure of 9c-A was similar to those
of 9a and 9b. On the basis of the IR and NMR spectra
the second isomer was assigned the structure 9c-B (see
eq 4).
Complex 5 did not react with iodine, even when iodine
was employed in large excess. In contrast, addition of
bromine to solutions of 5 in CDCl3 gave 9b by oxidative
decarbonylation, in addition to some byproduct. The
formation of byproducts has not been observed in the
reaction of 5 with SnBr4. Since oxidative decarbonyla-
tion of 5 by SnBr4 also leads to somewhat higher yields
of 9b, it is more convenient to prepare 9b from 5 and
SnBr4.
(13) See, e.g.: Wardell, J . L. In Encyclopedia of Inorganic Chemistry;
King, R. B., Ed.; Wiley: Chichester, U.K., 1994; Vol. 8, pp 4159-4171.

A New Class of Fischer Type Carbene Complexes
Organometallics, Vol. 20, No. 19, 2001 4043
Sch em e 1
investigated the reaction of SnBr4 with the para-MeO-
substituted complex cis-[(CO)4(PnBu3)WdC(OMe)C6H4-
OMe-p] (11)9b in which the PPh3 has been replaced by
the less sterically demanding and more strongly electron
donating tri-n-butylphosphine. To prevent chelation, the
o-MeO group has been shifted into the para position.
In CDCl3 11 quickly reacted with SnBr4 in excess. From
the 1H and the 31P NMR spectra of the reaction solution
the formation of three isomers (12-A, 12-B, and 12-B′,
12-B and 12-B′ being conformers) in a ratio of ap-
proximately 2:1:1 could be deduced (eq 6).
(close to 1:1) was independent of whether the cis or the
trans isomer of the starting complex 13 was employed.
From these observations it follows that chelation is
not required for the formation of stable carbene tung-
sten(II) complexes. In addition to the two σ- and
π-donating halides one strongly electron donating coli-
gand such as PnBu3 suffices to stabilize carbene tung-
sten(II) complexes, provided there is no steric congestion
at the W(II) center.
The failure to synthesize a dihalogeno carbene-C,O
chelate tricyclohexylphosphine complex is thus readily
explained by steric congestion at the metal. When SnBr4
was added to a solution of the tricyclohexylphosphine
complex 8 there was no indication for the formation of
dihalogeno carbene-C,O chelate tungsten(II) complexes.
The reaction mixture turned dark upon addition of the
halide. Although a product could not be isolated, the IR
spectrum of the reaction mixture indicated the forma-
tion of a carbyne complex, presumably trans-[Br(CO)4-
WtC(C6H4OMe-o)].14
Rea ction of Dih a logen o Ca r ben e-C,O Ch ela te
Com p lexes w ith P h osp h in es. The synthesis of car-
bene-C,O chelate phosphine complexes of molybdenum-
(II) from carbene carbonyl phosphine molybdenum(0)
complexes and SnX4 is not feasible, due to the lability
of the molybdenum(0) starting complexes. An alterna-
tive approach to carbene-C,O chelate phosphine metal
complexes would be substitution of PR3 for a CO ligand
in the readily available carbene-C,O chelate metal(II)
complexes 3 and 4 (see eq 1). It has already been shown
The para-unsubstituted complex [(CO)4(PnBu3)-
WdC(OMe)Ph] (13) reacted similarly with SnBr4. Two
isomers of [Br2(CO)3(PnBu3)WdC(OMe)Ph] (14-A and
14-B) were identified by NMR spectroscopy. The ratio
(14) (a) Fischer, E. O.; Kreis, G.; Kreiter, C. G.; Mu¨ller, J .; Huttner,
G.; Lorenz, H. Angew. Chem. 1973, 85, 619-621; Angew. Chem., Int.
Ed. Engl. 1973, 12, 564-565.

4044 Organometallics, Vol. 20, No. 19, 2001
Stumpf et al.
earlier that in tungsten halogeno carbonyl complexes
one carbonyl ligand can be replaced by a phosphine
ligand through substitution.15 We therefore briefly
studied the reactions of carbene-C,O chelate tricarbonyl
dihalogeno complexes with phosphine.
compounds were rather low (40 and 56%, respectively)
due to their poor solubility. The complexes 16 and 17
were characterized by their IR and 1H NMR spectra,
which supported the structural proposal shown in eq 8.
Very likely, steric congestion prevented the formation
of halogeno carbene-C,O chelate dicarbonyl triphen-
ylphosphine complexes.
The tungsten complexes 3a and 3b reacted with
equimolar amounts of PMe3 cleanly to give the dihalo-
geno carbene-C,O chelate phosphine complexes 9a and
9b, respectively, described above. This substitution
approach could be extended to the corresponding mo-
lybdenum complexes. In NMR experiments it was
possible to show that upon treatment of 4a and 4b with
equimolar amounts of PMe3 the molybdenum complexes
15a and 15b were readily and quantitatively formed.
They were identified by their IR and NMR spectra,
which are very similar to those of 9a and 9b. Therefore,
similar structures were assigned to all four complexes
(eq 7).
Discu ssion
Rea ction s. Our experiments demonstrate that an
opening of the chelate ring in the carbene-C,O chelate
tungsten(0) complex 1 by nucleophiles is possible and
that the resulting carbene phosphine complexes are
stable and isolable species. The Mo analogue of 1 reacts
similarly with PPh3 at -40 °C; however, the ring
opening is reversible. The species isolated at room
temperature is not, as with 1, a carbene tetracarbonyl
phosphine complex but rather a product derived from
substitution of PPh3 for a CO ligand. The corresponding
tungsten complex [(CO)3(PPh3)WdC(OMe)C6H4OMe-o]
is formed when solutions of 7 are irradiated in THF at
-30 °C.8
The opening of the chelate in 1 requires strong
nucleophiles such as phosphines. At room temperature
complex 1 did not react with simple olefins such as
1-pentene, ethyl vinyl ether, vinyl acetate, malonic acid
and fumaric acid dimethyl esters, and 2-methyl-1,3-
butadiene.8 Without a cocatalyst complex 1 is also
inactive in the ring-opening metathesis polymerization
of strained olefins such as norbornadiene, norbornene,
and cyclopentene. When terminal alkynes are employed,
polymerization of the alkyne is observed, albeit in low
yield.8
As expected on the basis of these previous results, the
reaction of the chloro carbene-C,O chelate complex 3a
with PnBu3 also afforded a single isomer (10a ), whereas
the two isomeric products 10b-A and 10b-B in a ratio
of 2.3:1 (see Scheme 1) were formed when the bromo
carbene-C,O chelate complex 3b was treated with
PnBu3. Additional products were not observed. The ratio
10b-A:10b-B was slightly larger as compared to that
obtained by the oxidative decarbonylation route (Scheme
1).
The substitution pathway even made halogeno car-
bene triphenylphosphine complexes accessible. How-
ever, the phosphine did not displace a CO ligand but
rather the coordinating MeO group of the chelate. Thus,
the reactions of 3b and 4b with PPh3 gave the non-
chelating carbene phosphine complexes 16 and 17,
respectively (eq 8). The substitution of PPh3 for a
The nonchelated carbene phosphine complexes 5 and
6 readily react with tin(IV) halides by oxidative decar-
bonylation to give halogeno carbene-C,O chelate carbo-
nyl phosphine tungsten(II) complexes. Very likely the
reaction proceeds in two steps: either (a) oxidation of
the complex and replacement of one CO ligand by the
halides is followed by dissociation of another W-CO
bond and chelation or (b) loss of one CO ligand from 5
or 6 and chelation is followed by oxidation of the
resulting carbene-C,O chelate complex and substitution
of two halides for another CO ligand. Neither interme-
diate expected in these pathways could be detected
spectroscopically. However, since the electron density
and thus the W-CO back-bonding is reduced by oxida-
tion of the metal, pathway a seems more likely. The
assumption is supported by the stability of 5 and 6 at
ambient temperature in the absence of SnX4. These
results indicate that chelation is not a prerequisite for
an oxidative decarbonylation of 5 and 6 as long as the
predominantly metal-centered HOMO is sufficiently
high in energy. This can also be achieved by substitution
of a phosphine for the chelating OMe group. The
conclusion is supported by the successful oxidative
decarbonylation of complex 11. In the resulting carbene
phosphine tungsten(II) complex the para position of the
aryl-OMe substituent prevents chelation through OMe.
Indeed, even the aryl-unsubstituted methoxy(phenyl)-
carbene complex 13 is transferred into a W(II) carbene
complex by SnBr4. Apart from these complexes the
oxidative decarbonylation of group 6 carbene carbonyl
complexes has also been achieved by reaction of Br2 with
carbonyl ligand could not be observed. The products 16
and 17 were isolated from the reaction mixture as red
powders. An excess of PPh3 was necessary to drive the
reaction of the molybdenum complex to a reasonable
conversion. Nevertheless, the yields of isolated pure
(15) Umland, P.; Vahrenkamp, H. Chem. Ber. 1982, 115, 3565-3579.

A New Class of Fischer Type Carbene Complexes
Organometallics, Vol. 20, No. 19, 2001 4045
(meso) isomers are conceivable; for the dihalogeno
carbene-C,O chelate phosphine complexes [MX2Y2Z-
(L-L′)] these numbers increase to 148 pairs of enanti-
omers and 4 achiral isomers.18 The dihalogeno carbene-
C,O chelate tricarbonyl complexes 3a -c and 4a ,c (eq
1) and the dihalogeno carbene-C,O chelate dicarbonyl
phosphine complexes 9a -c, 10a -c, and 15a ,b (eqs 3-5
and 7 and Scheme 1) are formed either as a single
isomer or at least as mixtures of two isomers. Therefore,
strong steric and/or electronic effects must be operating,
leading to the preference for only a few structures.
Hoffmann et al. gave a detailed account on seven-
coordination based on EHMO calculations.19 They iden-
tified the electronic factor that determines the prefer-
ence of certain ligands for the different positions and
provided some low-energy pathways for polytopal rear-
rangements. In d4 complexes with co geometry, π-ac-
ceptor ligands prefer the cf over the c and the uf
positions, while π donor ligands prefer the uf over the c
and the cf positions. These predictions are partially
fulfilled by the structures of 3a 6 and 9b, as determined
by X-ray structural analyses. In both complexes two
π-donor ligands (Cl, Br) occupy a uf position, whereas
the third π-donor (OMe) is in a cf position. The remain-
ing two cf positions are occupied, as expected by the
strong π-acceptors CO. In both complexes the π-acceptor
Ccarbene is in the c position, thus forcing the OMe
substituent into the “unfavorable” cf position. In con-
trast to the case for 3a , the structure of the correspond-
ing iodo carbene-C,O chelate complex 3c agrees with
the predictions: all π-donor ligands (I- and OMe) are
in the uf position. The π-acceptor ligands are at either
the cf or the c position, Ccarbene occupying a cf position.
The A structure of 9c and 10b corresponds to that of
9b. The detailed geometry of the B type isomer is at
present unknown. Several structures derived from A are
conceivable. From the 13C NMR spectra it follows that
both CO ligands are inequivalent. Except for Ccarbene and
OMe, all mutual changes of the positions of two unlike
ligands in these complexes will render the two CO
ligand inequivalent. Since the A and the B isomers do
not interconvert and since the product ratio A:B de-
pends on the reaction conditions, both isomers are
formed in parallel pathways. Considering this and the
increasing preference for isomer B in the series 9a -c
and 10a -c, the structure of isomer B very likely is
derived from that of 3c20 by substitution of PR3 for a
CO ligand: both halides and the OMe substituent
occupy the uf position and at least one CO ligand and
Ccarbene the cf position. PR3 then resides either in the
remaining cf position or in the c position. The increasing
preference for isomer B in the series 9a -c and 10a -c
is presumably due to increasing steric demand of the
different ligand (Cl, Br, I and PMe3, PnBu3).
F igu r e 1.
some Lappert type carbene complexes in which the
electron density at the metal is increased by the cyclic
strongly electron donating bis(amino)carbene ligands
dC[N(R)-CH2-]2.16
At first glance, the formation of the dibromo carbene-
C,O chelate tricarbonyl complex instead of the expected
dibromo carbene-C,O chelate dicarbonyl phosphine tung-
sten(II) complex 3b in the reaction of the triphenylphos-
phine complex 7 with SnBr4 is surprising. Very likely,
the W-PPh3 bond in [Br2(CO)3(PR3)WdC(OMe)C6H4-
OMe-o] (16b), initially formed in the reaction of 7 with
PPh3, is labile, as is also observed with [(CO)4(PR3)-
ModC(OMe)C6H4OMe-o].12 As a consequence, PPh3
readily and reversibly dissociates from the metal. Trap-
ping of free PPh3 by SnBr4 presumably is more rapid
than readdition of PPh3 to the metal. This interpretation
agrees well with the requirement of 2 equiv of SnBr4
for the complete conversion of 7 into 3b. In the absence
of the trapping agent SnBr4 complex 16b is stable and
is readily accessible by opening of the chelate ring in
3b by PPh3 (see eq 8).
The failure to synthesize the initially expected di-
bromo carbene-C,O chelate dicarbonyl triphenylphos-
phine tungsten(II) complex as well as the corresponding
tricyclohexylphosphine tungsten(II) complex is very
likely due to steric congestion at the heptacoordinated
tungsten atom. Bulky phosphines with large cone angles
seem to prevent the formation of isolable heptacoordi-
nated carbene-C,O chelate phosphine metal(II) com-
plexes. This also explains why complex 3b and the
related molybdenum(II) complex react with PPh3 by ring
opening to form heptacoordinated nonchelated dibromo
tricarbonyl carbene phosphine metal(II) complexes,
whereas in the reaction of 3b with PMe3 dibromo
carbene-C,O chelate dicarbonyl phosphine metal(II)
complexes are obtained.
Str u ctu r e. At least three different coordination
polyhedra are conceivable for seven-coordinate com-
plexes, the capped octahedron (co), the capped trigonal
prism (ctp), and the pentagonal bipyramid (pbp). The
heptacoordinated carbene complexes 3a 6 and 9b are
best described by the co geometry. In the co geometry
(Figure 1), three different positions can be identified:
the unique capping position (c, 7), the 3-fold capped face
(cf, 4-6), and the 3-fold uncapped face (uf, 1-3).
Haigh and Baker have used Po´lya’s theorem17 to
calculate the number of possible isomers in co structures
for any combination of ligands, including symmetrical
(L-L) and unsymmetrical (L-L′) chelating ligands. For
the dihalogeno chelate carbene complexes of the type
[MX2Y3(L-L′)], 48 pairs of enantiomers and 4 achiral
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All operations were carried out
under nitrogen by using conventional Schlenk techniques.
Solvents were dried by refluxing over sodium/benzophenone
ketyl or CaH2 and were freshly distilled prior to use. The silica
(18) Haigh, C. W.; Baker, P. K. Polyhedron 1994, 13, 417-433.
(19) Hoffmann, R.; Beier, B. F.; Muetterties, E. L.; Rossi, A. R. Inorg.
Chem. 1977, 16, 511-522.
(16) Lappert, M. F.; Pye, P. L. J . Chem. Soc., Dalton Trans. 1977,
1283-1291.
(17) Po´lya, G. Acta Mathematica 1937, 68, 145-254.
(20) Stumpf, R. Dissertation, Universita¨t Konstanz, 1997.

4046 Organometallics, Vol. 20, No. 19, 2001
Stumpf et al.
gel used for chromatography (J . T. Baker, silica gel for flash
chromatography) was dried in vacuo and saturated with
nitrogen. The yields refer to analytically pure compounds and
were not optimized. The complexes 1,21 3a ,6 3b,6 4a ,6 4b,6 and
137 were prepared according to literature procedures. Complex
11 was prepared as described in ref 4 for 13. Acetyl bromide
was purchased from Fluka. Instruments used were as fol-
lows: IR, FT-IR spectrophotometer, Bio-Rad; 1H NMR and 13C
NMR, Bruker WM 250, Bruker AC 250, J EOL J NX 400; 31P
NMR, J EOL J NX 400; MS, Finnigan MAT 312 (EI) or
Finnigan MAT 312/AMD5000 (FAB). All spectra were recorded
at room temperature. Unless specifically mentioned, chemical
shifts are reported relative to TMS (1H NMR spectra), to the
residual solvent peaks (13C NMR spectra; CDCl3 δ 77.0, CD2-
Cl2 δ 53.8, acetone-d6 δ 206.6), or to external H3PO4 (31P NMR
spectra).
solvent gave a light red powder. Recrystallization from pen-
tane/CH2Cl2 (1:1) afforded 1.2 g (87%) of complex 7 as light-
red crystals. Mp: 125 °C. IR (pentane): ν(CO) 2025 m, 1936
vs, 1926 sh, 1911 vs, 1897 s cm-1. 1H NMR (CDCl3): δ 3.47 (s,
3H, aryl OMe), 3.98 (s, 3H, carbene OMe), 6.34 (dd, J ) 7.3
and 1.5 Hz, 1H, C6H4), 6.72 (m, 2H, C6H4), 7.12 (ddd, J ) 8.1,
7.6, and 1.5 Hz, 1H, C6H4), 7.4 (m, 15H, Ph). 13C NMR (CDCl3,
-10 °C): δ 54.7 (s, aryl OMe), 65.4 (s, carbene OMe), 110.2,
119.4, 123.0, 128.2, 128.7, 129.8, 133.2, 135.7, 136.3, 149.6 (aryl
C), 202.5 (d, J PC ) 28.8 Hz, CO), 205.8 (d, J PC ) 10.2 Hz, CO),
212.8 (d, J PC ) 10.2 Hz, CO), 320.2 (d, J PC ) 7.1 Hz, carbene
C). 31P NMR (CDCl3): δ -3.15 (s and d, J PW ) 230 Hz). MS/
EI (FAB): m/z (%) 708 (17) [M+], 680 (16) [M+ - CO], 652 (100)
[M+ - 2CO], 624 (57) [M+ - 3CO], 596 (47) [M+ - 4CO], 581
(56) [M+ - 4CO - CH3], 445 (57) [M+ - PPh3]. Anal. Calcd for
C
31H25O6PW (708.4): C, 52.56; H, 3.56. Found: C, 52.59; H,
3.61.
Tet r a ca r b on yl[m et h oxy(o-m et h oxyp h en yl)ca r b en e]-
cis-Te t r a ca r b on yl[m e t h oxy(o-m e t h oxyp h e n yl)ca r -
ben e](tr im eth ylp h osp h in e)tu n gsten (0) (5). At room tem-
perature, 140 µL (0.10 g, 1.4 mmol) of PMe3 was added by
syringe to a solution of 1 (0.61 g, 1.3 mmol) in 25 mL of CH2-
Cl2. The solution turned red. The solvent was removed in
vacuo, and the residue was chromatographed on silica at -30
°C. With pentane/CH2Cl2 (ratio decreasing from 5:1 to 1:5) a
red-orange fraction was eluted. After removal of the solvent,
the residue solidified to give a red crystalline mass, yield 0.51
g (0.98 mmol, 72%). Mp: 71 °C. IR (pentane): ν(CO) 2023 m,
(tr icycloh exylp h osp h in e)tu n gsten (0) (8). The preparation
and purification of 8 were carried out analogously to 5.
Complex 8 was obtained as a mixture of the cis and trans
isomers (ratio: ca. 4.5:1) as a red-orange powder. Yield: 67%.
IR (pentane): ν(CO) 2018 m, 1918 m, 1910 vs, 1899 s cm-1
.
1H NMR (CDCl3): δ 1.22-1.94 (m, 33H, PCy3), 3.77 and 3.78
(2 s, 3H, aryl OMe, cis and trans), 3.85 and 4.30 (2 s, 3H,
carbene OMe, cis and trans), 6.80-7.26 (m, 4H, aryl H). 13C
NMR (CDCl3): δ 26.4, 27.6, 27.7, 27.9, 30.1, 37.2, 37.4 (s,
PCy3), 54.9 and 55.3 (2 s, aryl OMe), 63.3 and 65.1 (2 s, carbene
OMe), 109.9, 110.8, 119.7, 120.0, 122.2, 125.7, 127.6, 129.0,
143.1, 147.0, 148.7, 150.2 (12s, aryl C), 204.1 (d, J PC ) 6.7 Hz,
CO), 205.2 (d, J PC ) 6.0 Hz, CO), 206.8 (d, J PC ) 21 Hz, CO),
214.5 (d, J PC ) 7.8 Hz, CO), carbene C not found. 31P NMR
(CDCl3): δ 34.2 (s and d, J PW ) 230 Hz) and 30.8 (s and d,
1924 s, 1907 vs, 1892 m cm-1 1H NMR (CDCl3): δ 1.55 (d,
.
J PH ) 7.9 Hz, 9H, PMe3), 3.77 (s, 3H, aryl OMe), 4.12 (s, 3H,
carbene OMe), 6.82-6.85 (m, 1H, aryl H), 6.97-6.99 (m, 2H,
aryl H), 7.17-7.24 (m, 1H, aryl H). 13C NMR (CDCl3): δ 20.9
(d, J PC ) 27.6 Hz, PMe3), 54.8 (s, aryl OMe), 65.2 (s, carbene
OMe), 110.4, 120.1, 124.6, 128.8, 145.2, 149.2 (6s, aryl C), 202.5
(d, J PC ) 7.7 Hz, 2CO), 207.4 (d, J PC ) 19.3 Hz, CO), 213.2 (d,
J PC ) 7.8 Hz, CO), 320.9 (d, J PC ) 7.4 Hz, carbene C). 31P NMR
(CDCl3): δ -32.9 (s and d, J PW ) 230 Hz). MS/EI (70 eV, 100
°C); m/z (%) 522 (40) [M+], 494 (20) [M+ - CO], 466 (35) [M+
- 2CO], 438 (100) [M+ - 3CO], 410 (80) [M+ - 4CO]. Anal.
Calcd for C16H19O6PW (522.2): C, 36.80; H, 3.67. Found: C,
36.62; H, 3.71.
J
PW not found). Anal. Calcd for C31H43O6PW (726.5): C, 51.25;
H, 5.97. Found: C, 51.0; H, 5.90.
Dica r b on yld ich lor o[m et h oxy(o-m et h oxyp h en yl)ca r -
ben e-K2C,O](tr im eth ylp h osp h in e)tu n gsten (II) (9a ). A 7.0
mL portion (1.0 mmol) of a dilute solution of PMe3 (0.145 M
in CH2Cl2) was added to a vigorously stirred solution of 0.49
g (1.0 mmol) of 3a in 25 mL of CH2Cl2. The mixture was stirred
for 10 min. The volume of the solution was then reduced in
vacuo to ca. 10 mL. A layer of 10 mL of pentane was placed
on top of this solution. When the mixture was cooled to -30
°C, small red crystals formed overnight. Yield: 0.34 g (0.63
mmol, 62%). Mp: 140 °C dec. IR (CH2Cl2): ν(CO) 1970 m, 1888
vs cm-1. 1H NMR (CDCl3): δ 1.94 (d, J PH ) 10.7 Hz, 9H, PMe3),
4.39 (s, 3H, carbene OMe), 4.84 (s, 3H, aryl OMe), 7.19-7.35
(m, 2H, aryl H), 7.50-7.53 (m, 1H, aryl H), 7.76-7.80 (m, 1H,
aryl H). 13C NMR (CD2Cl2, -80 °C): δ 15.3 (d, J PC ) 37.4 Hz,
PMe3), 61.2 (s, carbene OMe), 62.4 (s, aryl OMe), 110.4, 122.3,
122.7, 133.6, 133.8, 161.1 (6s, aryl C), 223.0 (d, J PC ) 21.8 Hz
and dd, J PC ) 20.0 Hz and J WC ) 133 Hz, CO), 274.0 (dd, J PC
) 13.2 Hz, J WC not found, carbene C). 31P NMR (CDCl3): δ
-10.4 (s and d, J PW ) 223 Hz). Anal. Calcd for C14H19Cl2O4-
PW (537.0): C, 31.31; H, 3.57. Found: C, 30.88; H, 3.60.
Dib r om od ica r b on yl[m et h oxy(o-m et h oxyp h en yl)ca r -
ben e-K2C,O](tr im eth ylp h osp h in e)tu n gsten (II) (9b). A 1.3
g amount (2.9 mmol) of SnBr4 was added in small portions to
a solution of 1.5 g (2.9 mmol) of 5 in 40 mL of CH2Cl2. The
solution was stirred for 30 min and then filtered through a
layer of Celite. The volume of the solution was reduced to 10
mL. A layer of 10 mL of pentane was placed on top of this
solution. When this mixture was cooled to -30 °C, red crystals
formed overnight. They were collected by filtration and dried
in vacuo. Yield: 1.5 g (2.4 mmol, 84%). Mp: 140 °C dec. IR
cis-Te t r a ca r b on yl[m e t h oxy(o-m e t h oxyp h e n yl)ca r -
ben e](tr i-n -bu tylp h osp h in e)tu n gsten (0) (6). Preparation
and purification of 6 were carried out analogously to 5.
However, to avoid separation problems on chromatography,
only 0.9 equiv of PnBu3 was used. Complex 6 was obtained as
a red powder (86%). Mp: 44 °C. IR (pentane): ν(CO) 2020 m,
1920 s, 1904 vs, 1890 m cm-1 1H NMR (CDCl3): δ 0.92 (t,
.
J HH ) 6.9 Hz, 9H, PBu3), 1.30-1.43 (m, 12H, PBu3), 1.70-
1.79 (m, 6H, PBu3), 3.77 (s, 3H, aryl OMe), 4.02 (s, 3H, carbene
OMe), 6.81-6.85 (m, 1H, aryl H), 6.97-7.02 (m, 2H, aryl H),
7.18-7.25 (m, 1H, aryl H). 13C NMR (CDCl3): δ 13.8 (s, PBu3),
24.3 (d, J PC ) 12.8 Hz, PBu3), 25.9 (s, PBu3), 29.1 (d, J PC
)
23.9 Hz, PBu3), 54.9 (s, aryl OMe), 64.3 (s, carbene OMe),
110.3, 120.0, 124.9, 128.8, 144.6, 149.1 (6 s, aryl C), 202.8 (d,
J PC ) 8.2 Hz, 2CO), 206.9 (d, J PC ) 19.3 Hz, CO), 213.6 (d,
J PC ) 8.1 Hz, CO), 321.2 (d, J PC ) 7.4 Hz, carbene C). 31P NMR
(CDCl3): δ -3.15 (s and d, J PW ) 230 Hz). MS/EI (70 eV, 100
°C): m/z (%) 648 (12) [M+], 620 (3) [M+ - CO], 592 (13) [M+
-
2CO], 564 (28) [M+ - 3CO], 536 (5) [M+ - 4CO], 446 (9) [M+
- PBu3], 362 (28) [M+ - PBu3 - 3CO], 202 (22) [PBu3+]. Anal.
Calcd for C25H37O6PW (648.4): C, 46.31; H, 5.75. Found: C,
46.02; H, 5.71.
cis-Te t r a ca r b on yl[m e t h oxy(o-m e t h oxyp h e n yl)ca r -
ben e](tr ip h en ylp h osp h in e)tu n gsten (0) (7). A 1.6 g portion
(6.1 mmol) of PPh3 was added to a solution of 1 (0.90 g, 2.0
mmol) in 50 mL of CH2Cl2. The initially brown solution turned
slowly red. The solvent was removed in vacuo after 30 min,
and the residue was chromatographed at -40 °C with pentane/
CH2Cl2 (ratio decreasing from 5:1 to 2:1). Removal of the
(CH2Cl2): ν(CO) 1972 m, 1892 vs cm-1 1H NMR (CDCl3): δ
.
2.00 (d, J PH ) 10.7 Hz, 9H, PMe3), 4.41 (s, 3H, carbene OMe),
4.95 (s, 3H, aryl OMe), 7.18-7.37 (m, 2H, aryl H), 7.37-7.58
(m, 1H, aryl H), 7.79-7.83 (m, 1H, aryl H). 13C NMR (CDCl3,
-20 °C): δ 17.6 (d, J PC ) 37.3 Hz, PMe3), 62.6 (s, carbene
OMe), 63.5 (s, aryl OMe), 110.8, 122.8, 123.2, 134.1, 134.2,
(21) Do¨tz, K. H.; Erben, H.-G.; Staudacher, W.; Harms, K.; Mu¨ller,
G.; Riede, J . J . Organomet. Chem. 1988, 355, 177-181.

A New Class of Fischer Type Carbene Complexes
Organometallics, Vol. 20, No. 19, 2001 4047
162.1 (6s, aryl C), 219.8 (d, J PC ) 20.9 Hz and dd, J PC ) 21.0
Hz and J WC ) 135 Hz, CO), 273.9 (dd, J PC ) 13.6 Hz, J WC not
found, carbene C). 13C NMR (CH2Cl2, -80 °C): δ 16.9 (d, J PC
) 38.3 Hz, PMe3), 62.5 (s, carbene OMe), 63.5 (s, aryl OMe),
10b-A and 10b-B (ratio: ca. 2:1). Yield: 0.33 g (0.44 mmol,
53%). IR (CH2Cl2): ν(CO) ∼1960 vs (A and B), 1890 vs (A),
1859 s cm-1 (B). 1H NMR (CDCl3): δ 0.77 and 0.96 (2 t, J PH
)
7.3 and 7.2 Hz, 9H, PBu3 of B and A), 1.10-1.78 (m, 12H,
PBu3 of A/B), 2.17-2.67 (m, 6H, PBu3 of A/B), 4.39 and 4.63
(2 s, 3H, carbene OMe of A and B), 4.68, 4.84, and 4.87 (3 s,
3H, aryl OMe of A, B, and B′), 7.09-7.23 (m, 2H, aryl H of
A/B), 7.30-7.68 (m, 1H, aryl H of A/B), 7.78-7.88 (m, 1H, aryl
H of A/B). 13C NMR (CDCl3, -20 °C; only the resonances of
isomer A could be assigned unambiguously): δ 13.8 (s, PBu3),
24.0 (d, J PC ) 14.0 Hz, PBu3), 25.4 (d, J PC ) 4.6 Hz, PBu3),
26.3 (d, J PC ) 30 Hz, PBu3), 61.0 (s, carbene OMe), 63.1 (s,
aryl OMe), 111.5, 122.4, 124.7, 127.7, 135.3, 161.7 (6 s, aryl
C), 218.3 (d, J PC ) 18.8 Hz and d, J PC ) 19.1 Hz, 2CO), 272.7
(d, J PC ) 12.6 Hz, carbene C). 31P NMR (CDCl3): δ 15.2 (s and
d, J PW ) 230 Hz, A), 2.8 (s and d, J PW ) 223 Hz, B), 2.7 (s and
d, J PW ) 223 Hz, B′). Molecular formula: C23H37Br2O4PW
(752.2).
110.7, 122.8, 133.5, 134.0, 161.7 (5s, aryl C), 220.1 (d, J PC
)
20.7 Hz and dd, J PC ) 20.6 Hz and J WC ) 136 Hz, 2CO), 272.9
(dd, J PC ) 12.9 Hz, J WC not found, carbene C). 31P NMR
(CDCl3): δ -16.7 (s and d, J PW ) 215 Hz). Anal. Calcd for
C
14H19Br2O4PW (625.9): C, 26.86; H, 3.06. Found: C, 26.64;
H, 3.12.
Dicar bon yldiiodo[m eth oxy(o-m eth oxyph en yl)car ben e-
K2C,O](tr im eth ylp h osp h in e)tu n gsten (II) (9c-A/9c-B). A
0.88 g amount (1.4 mmol) of SnI4 was added in small portions
to a solution of 0.75 g (1.4 mmol) of 5 in 50 mL of CH2Cl2.
Immediately, a gas evolved and a voluminous yellow precipi-
tate formed. The solution was stirred for 10 min and then
filtered through a layer of Celite. The volume of the red
solution was reduced to ca. 5-10 mL. A layer of 10 mL of
pentane was placed on top of this solution. When the mixture
was cooled to -30 °C, red crystals formed overnight. They were
collected by filtration and dried in vacuo. Yield: 0.81 g (1.1
mmol, 79%). The red crystalline material consisted of both
isomers 9c-A and 9c-B (ratio: ca. 1:2). IR (CH2Cl2): ν(CO)
Dicar bon yldiiodo[m eth oxy(o-m eth oxyph en yl)car ben e-
K2C,O](tr i-n -bu tylp h osp h in e)tu n gsten (II) (10c). The syn-
thesis of 10c from 0.65 g (1.0 mmol) of 6 in 50 mL of CH2Cl2
and SnI4 (0.63 g, 1.0 mmol) was carried out analogously to
that of 9c-A/9c-B from 5 and SnI4. Yield: 0.79 g (0.93 mmol,
93%) of red crystals. Mp: 185 °C dec. IR (CH2Cl2): ν(CO) 1954
vs, 1863 s cm-1. 1H NMR (CDCl3): δ 0.78 (t, J PH ) 7.3 Hz, 9H,
PBu3), 0.81-1.35 (2m, 12H, PBu3), 1.63-1.81 (m, 3H, PBu3),
2.36-2.51 (m, 3H, PBu3), 4.51 (s, 3H, carbene OMe), 4.94 (s,
3H, aryl OMe), 7.21-7.34 (m, 2H, aryl H), 7.51-7.58 (m, 1H,
aryl H), 7.71-7.75 (m, 1H, aryl H). 13C NMR (CDCl3, -20 °C):
1964 vs (A and B), 1896 m (A), 1865 s (B) cm-1 1H NMR
.
(CDCl3): δ 1.57 and 2.09 (2 d, J PH ) 9.7 and 10.3 Hz, 9H, PMe3
of B and A), 4.43 and 4.53 (2 s, 3H, carbene OMe of A and B),
4.95 and 5.08 (2 s, 3H, aryl OMe of B and A), 7.18-7.40 (m,
2H, aryl H of A/B), 7.56-7.62 (m, 1H, aryl H of A/B), 7.72-
7.76 (m, 1H, aryl H of A/B). 13C NMR (CDCl3, -20 °C): δ 16.3
(d, J PC ) 35.4 Hz, PMe3 of B) 20.1 (d, J PC ) 39 Hz, PMe3 of A),
62.6 (s, carbene OMe of A), 66.2 (s, carbene OMe of B), 67.3
(s, aryl OMe of B), 67.9 (s, aryl OMe of A), 111.2, 122.8, 123.2,
134.4, 135.5, 163.1 (6 s, aryl C of A), 113.3, 123.1, 124.7, 133.0,
135.5, 163.1 (6 s, aryl C of B) 215.5 (d, J PC ) 20.7, 2CO of A),
227.9 (d, J PC ) 7.3, CO of B), 245.6 (d, J PC ) 30.5, CO of B),
273.2 (d, J PC ) 13.4, carbene C of A), 301.0 (d, J PC ) 6.1 Hz,
δ 13.8 (s, PBu3), 24.3 (d, J PC ) 14.1 Hz, PBu3), 24.7 (d, J PC
)
28.2 Hz, PBu3), 25.3 (d, J PC ) 2.6 Hz, PBu3), 66.0 (s, carbene
OMe), 67.5 (s, aryl OMe), 113.4, 122.7, 124.6, 133.6, 135.1,
163.8 (6 s, aryl C), 229.3 (d, J PC ) 6.4 Hz, CO), 245.1 (d, J PC
) 29.5 Hz, CO), 301.4 (d, J PC ) 7.1 Hz, carbene C). 31P NMR
(CDCl3): δ 7.74 (s and d, J PW ) 228 Hz). Anal. Calcd for
C
23H37I2O4PW (846.2): C, 32.65; H, 4.41. Found: C, 33.11; H,
4.62.
Dibr om otr ica r bon yl[m eth oxy(p-m eth oxyp h en yl)ca r -
carbene C of B). 31P NMR (CDCl3): δ -11.45 (s and d, J PW
)
237 Hz, B) and δ -23.65 (s and d, J PW ) 210 Hz, A). Anal.
Calcd for C14H19I2O4PW (719.0): C, 23.36; H, 2.66. Found: C,
23.56; H, 2.61.
ben e](tr i-n -bu tylp h osp h in e)tu n gsten (II) (12-A/12-B/12-
B′). In an NMR tube, ca. 50 mg of 11 was dissolved in CDCl3.
1
SnBr4 was added in excess, and the H and 31P NMR spectra
Dica r b on yld ich lor o[m et h oxy(o-m et h oxyp h en yl)ca r -
ben e-K2C,O](tr i-n -bu tylp h osp h in e)tu n gsten (II) (10a ). A
freshly prepared diluted solution of 0.18 g (0.23 mL, 0.89
mmol) of tri-n-butylphosphine in 10 mL of CH2Cl2 was slowly
added to a vigorously stirred solution of 0.45 g (0.9 mmol) of
3a in 25 mL of CH2Cl2. The solution was stirred for 10 min
and then filtered through a layer of Celite. The volume of the
solution was reduced to 10 mL. A layer of 10 mL of pentane
was placed on top of this solution. When the mixture was
cooled to -30 °C, red crystals of 10a formed overnight. They
were collected by filtration and dried in vacuo. Yield: 0.48 g
(0.72 mmol, 80%). Mp: 177 °C dec. IR (CH2Cl2): ν(CO) 1968
were taken at room temperature. The solvent was removed in
vacuo. The residue was dissolved in CH2Cl2 and the IR
spectrum recorded. IR (CH2Cl2): ν(CO) 2011 m sh, 1986 s, 1946
1
m, 1900 vs cm-1. H NMR (CDCl3): δ 0.94 and 0.99 (2 t, J HH
) 7.0 Hz and J HH ) 7.1 Hz, 9H, PBu3 of B and A), 1.25-1.67
(m, 12H, PBu3 of A and B), 2.07-2.65 (m, 6H, PBu3 of A and
B), 3.86, 3.93, and 3.95 (3 s, 3H, aryl OMe of A, B and B′),
4.20 and 4.26 (2 s, 3H, carbene OMe of A and B) 6.91 (“d”,
J HH ) 8.8 Hz, 1H, aryl H of A), 7.06 (“d”, J HH ) 9.0 Hz, 1H,
aryl H of B), 7.61 (“d”, J HH ) 9.0 Hz, 1H, aryl H of B), 7.79
(“d”, J HH ) 8.8 Hz, 1H, aryl H of A). 31P NMR (CDCl3): δ 14.8
(s and d, J WP ) 243 Hz, A), 10.1 (s and d, J WP ) 223 Hz, B),
m, 1886 vs cm-1 1H NMR (CDCl3): δ 0.97 (t, J HH ) 7.2 Hz,
.
9.6 (s and d, J WP ) 223 Hz, B′). Molecular formula: C24H37
Br2O5PW (780.2).
-
9H, PBu3), 1.39-1.58 (m, 6H, PBu3), 1.61-1.72 (m, 6H, PBu3),
2.22-2.32 (m, 6H, PBu3), 4.39 (s, 3H, carbene OMe), 4.75 (s,
3H, aryl OMe), 7.14-7.21 (m, 1H, aryl H), 7.26-7.29 (m, 1H,
aryl H), 7.46-7.53 (m, 1H, aryl H), 7.73-7.77 (m, 1H, aryl
H). - 13C NMR (CDCl3): δ 13.6 (s, PBu3), 24.1 (d, J PC ) 13.4
Hz, PBu3), 25.4 (d, J PC ) 10.4 Hz, PBu3), 25.6 (d, J PC ) 14.6
Hz, PBu3), 60.5 (s, carbene OMe), 62.5 (s, aryl OMe), 110.6,
Dibr om otr ica r bon yl[m eth oxy(p h en yl)ca r ben e](tr i-n -
bu tylp h osp h in e)tu n gsten (II) (14-A/14-B). The generation
of 14-A/14-B in CDCl3 from ca. 50 mg of 13 and SnBr4 and
the spectroscopic investigations were carried out analogously
to those of 12-A/12-B/12-B′. The results were independent of
whether the cis or the trans isomer of the starting complex
was used. IR (CH2Cl2): ν(CO) 2026 w, 2016 w, 1987 m, 1940
122.8, 123.8, 133.2, 133.9, 161.5 (6 s, aryl C), 222.1 (d, J PC
)
18.8 Hz and dd, J PC ) 18.8 Hz and J WC ) 133 Hz, 2CO), 321.2
(d, J PC ) 12.5 Hz, carbene C). 31P NMR (CDCl3): δ 2.74 (s and
d, J PW ) 223 Hz). Anal. Calcd for C23H37Cl2O4PW (663.3): C,
41.65; H, 5.62. Found: C, 42.05; H, 5.85.
s, 1902 vs cm-1. 1H NMR (CDCl3): δ 0.95 and 0.99 (2 t, J HH
)
7.0 Hz and J HH ) 7.1 Hz, 9H, PBu3 of B and A), 1.26-1.65
(m, 12H, PBu3 of A and B), 2.07-2.66 (m, 6H, PBu3 of A and
B), 4.20 and 4.28 (2 s, 3H, carbene OMe of A and B), 7.38-
7.44 (m, 1H, aryl H of A and B), 7.55-7.66 (m, 3H, aryl H of
A and B), 7.75-7.83 (m, 1H, aryl H of A and B). Molecular
formula: C23H35Br2O4PW (750.2).
Dib r om od ica r b on yl[m et h oxy(o-m et h oxyp h en yl)ca r -
ben e-K2C,O](t r i-n -b u t ylp h osp h in e)t u n gst en (II) (10b-A/
10b-B). The synthesis of 10b-A/10b-B from 0.54 g (0.83 mmol)
of 6 and SnBr4 (17 mL, 0.83 mmol, 0.048 M in CH2Cl2) was
carried out analogously to that of 9c-A/9c-B from 5 and SnI4.
A red-brown powder was obtained, consisting of both isomers
Dica r b on yld ich lor o[m et h oxy(o-m et h oxyp h en yl)ca r -
ben e-K2C,O](tr im eth ylp h osp h in e)m olybd en u m (II) (15a ).

4048 Organometallics, Vol. 20, No. 19, 2001
Stumpf et al.
In an NMR tube, ca. 50 mg of 4a was dissolved in CD2Cl2.
PMe3 (1 equiv) was added, and the 1H and 31P NMR spectra
were taken at room temperature. The solvent was removed in
vacuo. The residue was dissolved in CH2Cl2 and the IR
a dark brick red powder. Mp: 135 °C dec. IR (CH2Cl2): ν(CO)
1964 s, 1870 s cm-1 1H NMR (CDCl3): δ 3.87 (s, 3H, aryl
.
OMe), 4.64 (s, 3H, carbene OMe), 6.96-7.00 (m, 1H, aryl H),
7.26-7.81 (m, 18H, aryl H and phenyl H). Due to poor
solubility 13C NMR and 31P NMR could not be obtained. Anal.
Calcd for C30H25Br2O5PW (840.2): C, 42.89; H, 3.10. Found:
C, 42.87; H, 3.15.
spectrum recorded. IR (CH2Cl2): ν(CO) 1981 m, 1907 vs cm-1
.
1H NMR (CD2Cl2): δ 1.84 (d, J PH ) 10.8 Hz, 9H, PMe3), 4.37
(s, 3H, carbene OMe), 4.67 (s, 3H, aryl OMe), 7.21-7.28 (m,
1H, aryl H), 7.34-7.37 (m, 1H, aryl H), 7.60-7.66 (m, 1H, aryl
H), 7.80-7.84 (m, 1H, aryl H). 13C NMR (CD2Cl2): δ 17.2 (d,
J PC ) 33.7 Hz, PMe3), 61.5 (s, carbene OMe), 63.4 (s, aryl OMe),
111.5, 123.1, 123.2, 132.5, 135.3, 162.8 (6 s, aryl C), 229.0 (d,
J PC ) 27.9 Hz), 289.5 (d, J PC ) 14.4 Hz). Molecular formula:
Dibr om otr ica r bon yl[m eth oxy(o-m eth oxyp h en yl)ca r -
ben e](tr ip h en ylp h osp h in e)m olybd en u m (II) (17). A 2.6 g
(5.4 mmol) portion of 4b was dissolved in 50 mL of CH2Cl2,
and PPh3 (2.8 g, 11 mmol) was added. After a short time, a
red precipitate formed. After 30 min at room temperature the
precipitate was collected either by filtering or by decanting
the solution. After drying in vacuo, the red powder was
analytically pure. Yield: 2.3 g, 3.0 mmol (56%). Mp: 117 °C
C
14H19Cl2MoO4P (449.1).
Dib r om od ica r b on yl[m et h oxy(o-m et h oxyp h en yl)ca r -
ben e-K2C,O](tr im eth ylp h osp h in e)m olybd en u m (II) (15b).
The generation of 15b in CD2Cl2 and the spectroscopic
investigations were carried out analogously to those of 15a .
dec. IR (CH2Cl2): ν(CO) 1976 vs, 1882 s cm-1 1H NMR
.
(CDCl3): δ 3.82 (s, 3H, aryl OMe), 4.64 (s, 3H, carbene OMe),
7.00-7.02 (m, 1H, aryl H), 7.28-7.45 (m, 16H, aryl H and
phenyl H), 7.60-7.64 (m, 1H, aryl H), 7.78-7.82 (m, 1H, aryl
H). 13C NMR and 31P NMR could not be recorded due to poor
solubility. Anal. Calcd for C30H25Br2MoO5P (752.3): C, 47.90;
H, 3.35. Found: C, 47.66; H, 3.38.
1
IR (CH2Cl2): ν(CO) 1981 m, 1910 vs cm-1. H NMR (CD2Cl2):
δ 1.83 (d, J PH ) 10.6 Hz, 9H, PMe3), 4.39 (s, 3H, carbene OMe),
4.77 (s, 3H, aryl OMe), 7.22-7.27 (m, 1H, aryl H), 7.35-7.38
(m, 1H, aryl H), 7.62-7.68 (m, 1H, aryl H), 7.83-7.86 (m, 1H,
aryl H). 13C NMR (CD2Cl2): δ 18.6 (d, J PC ) 34.9 Hz, PMe3),
63.4 (s, carbene OMe), 65.8 (s, aryl OMe), 111.6, 123.0, 123.1,
131.6, 135.3, 163.0 (6 s, aryl C), 226.0 (d, J PC ) 26.4 Hz, 2CO),
Ack n ow led gm en t. Support of this work by the
Deutsche Forschungsgemeinschaft and the Fonds der
Chemischen Industrie is gratefully acknowledged.
289.5 (d, carbene C). Molecular formula:
(538.0).
C14H19Br2MoO4P
Dibr om otr ica r bon yl[m eth oxy(o-m eth oxyp h en yl)ca r -
ben e](tr ip h en ylp h osp h in e)tu n gsten (II) (16). A 0.28 g
(0.48 mmol) portion of 3b was dissolved in 8 mL of CH2Cl2,
and PPh3 (0.13 g, 0.50 mmol) was added. The solution was
stirred at room temperature. After a red precipitate had
formed, the solution was cooled to -80 °C and the mixture
was decanted. The red residue was dried in vacuo and washed
twice with 10 mL of ether. Yield: 0.16 g (0.19 mmol, 40%) of
1
Su p p or tin g In for m a tion Ava ila ble: IR and H, 13C, and
31P NMR spectra of complexes 9c-A/9c-B, 10b, 10c, 12, 15a ,
15b, 16, and 17. This material is available free of charge via
the Internet at http://pubs.acs.org.
OM010429E