Infra-Red and Raman Spectra of Polyatomic Molecules

Abstract

12CO:,3CO/Ar. studies, although a few observations have been made in the gas phase.2 Within this series of monocarbonyls where the metal mass is relatively large, an excellent indication of the strength of the M-CO bond is the lowering of the CO vibration frequency from that in the free molecule where vqo = 2143 cm'1.3 (Gas to matrix shifts can be ignored here since they can be expected to amount to only a few reciprocal centimeters.4) The first measurements of this kind on NiCO were made by DeKock,5 and subsequently the CO stretching frequencies of most other members of the series have been measured, largely by Ozin's group.2 We have recently repeated the work on several transition-metal monocarbonyls in dilute matrices. For CrCO, the correct vco lies at 1977 cm'1 as Figure 1 clearly shows. ScCO yielded a broad weak band at 1950 cm"1 (shifted appropriately by 13CO substitution), but this assignment must be considered as tentative. We have also repeated the Mn + CO studies and observed the IR bands previously reported by Huber et al.6 But, as also concluded by those authors, none of the band combinations produced in dilute matrices of the metal atoms and mixtures of 12CO and I3CO is firmly assignable to MnCO. Huber et al. assign vco in MnCO at 1850 cm'1,7 but we could not justify such an assignment since it did not appear as a dominant band in our dilute matrices. In fact, we were concerned about our failure to identify such a band for MnCO and were led to consider the possibility that Mn atoms do not bond to CO. A plot of all of the presently available vco data is shown in Figure 2. If indeed Mn-CO is nonbonded, then vco = 2140 cm'1, and the familiar "double-humped" graph is now obtained which (2) VCO: Hanlan, L.; Huber, H.; Ozin, G. A. Inorg. Chem. 1976, 15, 2592-2597. Van Zee, R. J.; Bach, S. B. H.; Weltner, W" Jr. J. Phys. Chem. 1986, 90, 583-588. CrCO: ref 7, pp 135-137. MnCO: Huber, H.; Kundig, E. P.; Ozin, G. A.; Poe, A. Feigerle, C. S.; Lineberger, W. C. J. Am. Chem. Soc. 1982, 104, 5026-5031. CuCo: Huber, H.; Kuendig, E. P.; Moskovits, M.; Ozin, G. A. Figure 2. Plot of the CO stretching frequencies in the first-row transition metal monocarbonyl molecules MCO (circled points are tentative). Also shown is the variation of the energy of promotion corresponding to 4s23d""2-* 4s13d""1, where n is the number of valence electrons.8 is traditionally observed for the variation in properties across a row of transition metals. (This graph is a revised version of one given by Ozin and Vander Voet.7) This bonding behavior among the MCO molecules is then intuitively appealing. A more rational basis for this proposal is evident when the promotion energy s2-*· sd""1 of the first-row metals8 is also plotted (see Figure 2). A high promotion energy is interpreted as a high barrier to-bond formation and thereby a destabilization of the metal-CO bond. This does not explain the decreasing strength of bonding in going from FeCO to CuCO implied by the vco variation, but that trend can be attributed to repulsion arising from the increasingly filled 3d shell.9 We then suggest that theoretical work be attempted on MnCO and expect that it will be found, similar to CuCO, to be essentially nonbonded. The many and extensive calculations on NiCO presumably establish that it is *. Correspondingly, but less thoroughly treated, FeCO and CuCO are found to be 3 and 2 , respectively. The latter was corroborated via ESR by Kasai and Jones.10 Our attempts to observe FeCO in the same way have not been successful, indicating that the calculated ground state is incorrect or, if correct, the zero-field splitting in the triplet state is >3 cm"1.11 We have been unsuccessful in observing the other monocarbonyls via ESR except for VCO, which yielded a surprising , but quite definite, sextet ground state.12 The failure to observe ESR spectra for a (presumed linear) MCO molecule means that either (1) its ground state may be nonmagnetic (i.e., singlet, as calculated for NiCO), (2) if S > '/2, it may be orbitally degenerate (possibly this is the case for TiCO where a 3 ground state was predicted1), or (3) if 5 > */2, the zero-field splitting in the ground state is large (as possibly occurs for 3 FeCO). Acknowledgment. A referee of this paper has pointed out to us that there are at least three other factors, including overlap, hybridization, and exchange, which moderate the promotional effect upon bonding. He suggests that the correct promoted state of the metal here should be s°d", and he derives a curve in Figure 2 which more closely follows the variations in vco. However, even in his more refined theory, the high promotional energy of Mn is still the dominant factor, so our suggestion that MnCO is essentially nonbonded remains unaltered. (10) Kasai, P. H.; Jones, P. M.