1334
Can. J. Chem. Vol. 77, 1999
tion. The equilibrium depicted in eq. [4] is similar to the
as better donor ligands than the parent CpIr(CO)2. It should
also be pointed out that the acceptor fragments employed in
this investigation were neutral five-coordinate fragments and
it may be that complexes of CpIr(CO)2 with fragments posi-
tively charged or of lower coordination numbers (with fewer
steric interactions) might be more stable. The stable (η5-
C5Me5)[(PriO)3P](OC)IrRh(CO)[P(OPri)3](Cl) (5a) and (η-
C5H5)(OC)2RhPt(CO)(C6F5)2 (5b) complexes mentioned in
the Introduction are consistent with the latter suggestion.
equilibrium
that
we
have
observed
for
(OC)5OsRu(CO)3(SiCl3)(Br) (18).
The chloro analogue of 3, Cp(OC)2IrOs(CO)3(Cl)2, simi-
larly dissociated in CH2Cl2 at room temperature at approxi-
mately the same rate as 3, but there was apparently also
decomposition of [Os(µ-Cl)(CO)3(Cl)]2 so that after 24 h an
infrared spectrum of the reaction solution showed bands due
to CpIr(CO)2, trace amounts of Cp(OC)2IrOs(CO)3(Cl)2, and
[Os(µ-Cl)(CO)3(Cl)]2, plus unidentified products.
These preliminary results indicate that Os(CO)3(Br)2 is a
better acceptor than the iodo analogue, as might be expected
from the electronegativities of the two halogens. That the ac-
ceptor abilities of the Os(CO)3(X)2 (X = Cl, Br) fragments
toward CpIr(CO)2 are similar may be rationalized by invok-
ing more pπ donation from the halogen atoms to the dπ
orbitals of osmium for the chloro fragment compared to the
bromo analogue. This would decrease the Lewis acidity of
Os(CO)3(Cl)2 and offset the superior electron-withdrawing
ability of the Cl atoms. This is of course similar to the argu-
ments that have been used to explain why BF3 is a weaker
Lewis acid than BCl3 or BBr3 (23). Consistent with the view
that Cl pπ donation to Os dπ orbitals is significant is that the
Os—CO length trans to the Cl atom in 2 (1.85(1) Å) is con-
siderably shorter than the Os—C distances (1.94(1),
1.95(1) Å) to the carbonyl ligands that are in a mutually
trans configuration (Table 5).
The structure of 3 reveals (Fig. 5), once again, an
unbridged metal–metal bond. The length of the IrOs bond in
3 is 2.8540(7) Å, which is 0.016 Å shorter than the corre-
sponding length in 2. From a ground state viewpoint, there-
fore, the IrOs bonds in 2 and 3 are of comparable strength. It
has been proposed, however, that transition state stabiliza-
tion effects are important in determining the lability of da-
tive metal–metal bonds (4, 24). Besides the bond lengths
mentioned above, the IrOs distance in 3 may be compared to
Acknowledgments
We thank the Natural Sciences and Engineering Research
Council of Canada, the National University of Singapore,
and St. Mary’s University for financial support.
References
1. F.W.B Einstein, R.K Pomeroy, P. Rushman, and A.C. Willis. J.
Chem. Soc. Chem. Commun. 854 (1983).
2. H.B. Davis, F.W.B. Einstein, P.G. Glavina, T. Jones, R.K.
Pomeroy, and P. Rushman. Organometallics, 8, 1030 (1989).
3. J.A. Shipley, R.J. Batchelor, F.W.B. Einstein, and R.K.
Pomeroy. Organometallics, 10, 3620 (1991).
4. F. Jiang, J.L. Male, K. Biradha, W.K. Leong, R.K. Pomeroy,
and M.J. Zaworotko. Organometallics, 17, 5810 (1998).
5. (a) A.A. Del Paggio, E.L. Muetterties, D.M. Heinekey, V.W.
Day, and C.S. Day. Organometallics,5, 575 (1986); (b) R.
Usón, J. Forniés, P. Espinet, C. Fortuño, M. Tomas, and A.J.
Welch. J. Chem. Soc. Dalton Trans. 3005 (1988).
6. (a) F.H. Antwi-Nsiah, O. Oke, and M. Cowie.
Organometallics, 15, 1042 (1996); (b) T. Tanase, H. Toda, K.
Kobayashi, and Y. Yamamoto. Organometallics, 15, 5272
(1996); (c) M. Knorr, P. Braunstein, A. Tiripicchio, and F.
Ugozzoli. J. Organomet. Chem. 526, 105 (1996); (d) S.
Takemoto, S. Kuwata, Y. Nishibayashi, and M. Hidai. Inorg.
Chem. 37, 6428 (1998); (e) K. Mashima, A. Fukumoto, H.
Nakano, Y. Kaneda, K. Tani, and A. Nakamura. J. Am. Chem.
Soc. 120, 12 151 (1998).
7. F.W.B. Einstein, R.K. Pomeroy, P. Rushman, and A.C. Willis.
Organometallics, 4, 250 (1985).
8. F. Jiang, H.B. Davis, K. Biradha, R.K. Pomeroy, and M.J.
Zaworotko. Organometallics. Submitted.
9. (a) E.O. Fischer and K.S. Brenner. Z. Naturforsch. Teil B, 17,
774 (1962); (b) S.A. Garder, P.S. Andrews, and M.R. Rausch.
Inorg. Chem. 12,2396 (1973); (c) J.K. Hoyano and W.A.G.
Graham. Personal communication.
10. R.K. Pomeroy and K.S. Wijesekera. Inorg. Chem. 19, 3729
(1980).
the OsOs distance of 2.9104(6)
Å in the related
(OC)5OsOs(CO)3(Cl)2 recently prepared in this laboratory.
An interesting feature of the structure of 3 is that both the Br
atoms are cis to the IrOs bond rather than one Br atom trans
to the bond. Because the Br ligand is a π donor, the electron-
ically preferred position is expected to be cis to the dative
bond (see above). An alternative view is that the fac ar-
rangement of ligands on the acceptor moiety allows each
carbonyl ligand to be trans to a poor π acceptor ligand and
avoids two CO groups competing for the same electron den-
sity on the central Os atom. The Os—Br distances in 3
(2.574(2), 2.585(2)Å) are essentially the same as the Os—Br
lengths of 2.592(3) Å in Os3(CO)10[P(OMe3)3]2(Br)2 (25).
11. P. Rushman, G.N. van Buuren, M. Shiralian, and R.K.
Pomeroy. Organometallics, 2, 693 (1983).
12. L. Vancea and W.A.G. Graham. J. Organomet. Chem. 134, 219
(1977).
Conclusions
13. F. L’Eplattenier and F. Calderazzo. Inorg. Chem. 6, 2092
(1967).
14. J. L. Male. Ph.D. Thesis, Simon Fraser University. 1996.
15. G.M. Sheldrick. SHELXTL, Siemens, Madison, Wis. 1995.
16. A. Riesen, F.W.B. Einstein, A.K. Ma, R.K. Pomeroy, and J.A.
Shipley. Organometallics, 10, 3629 (1991).
17. F.W.B. Einstein, M.C. Jennings, R. Krentz, R.K. Pomeroy, P.
Rushman, and A.C Willis. Inorg. Chem. 26, 1341 (1987).
18. M.M. Fleming, R. K. Pomeroy, and P. Rushman. J.
Organomet. Chem. 278, C33 (1984).
The results of this study indicate that CpIr(CO)2 is a weak
ligand and only with 16e organometallic fragments that are
strong acceptors can complexes be isolated. Except for
Cp(OC)2IrOs(CO)3(GeCl3)(Cl), the complexes dissociate in
solution, which results in most cases in decomposition. The
ligating ability of CpIr(CO)2 therefore appears comparable
to that of Os(CO)5 (1, 2). Since Os(CO)4(PR3) compounds
are better ligands than Os(CO)5 (2a, 18), molecules of the
type CpIr(CO)(PR3) with small P ligands will probably act
© 1999 NRC Canada