FULL PAPER DOI: 10.1002/ejic.200600670 W(CO) 4 (diimine) Structure Revised – Correlating Structure to π* Back- Bonding Christodoulos Makedonas [a] and Christiana A. Mitsopoulou* [a] Keywords: Tungsten / Pi interactions / N ligands / Carbonyls / Ligand effects The bending of axial and equatorial carbonyls, one of the main structural features of the M(CO) 4 (diimine) compounds (M = Cr, Mo, and W), is elucidated in terms of the π* back- bonding theory by a frontier molecular orbital analysis. Em- ploying W(CO) 4 (phen) (phen = 1,10-phenanthroline) as a ref- erence compound has proved that the deviation from orthog- onality stabilizes the structure by a total amount of 2.95 kJ/ mol. Moreover, by an extended comparison of several W(CO) 4 (diimine) structures, the importance of the magnitude of the binding angle in determining the bonding in these complexes in comparison to the traditional approach of the W–C and C–O bond lengths is underlined. As a result, it is 1. Introduction The complexes of the type M(CO) 4 (diimine) (M = Cr, Mo, and W) play an important role in our understanding of the spectroscopic, photophysical, and photochemical be- havior of chromophores with low-lying metal-to-ligand (MLCT) charge-transfer excited states. [1–3] These com- pounds combine an electron-rich, low-valent, typically d 6 metal atom, which is stabilized by the presence of four car- bonyl groups along with the good π*-accepting α-diimine ligand. The observed lowest lying MLCT state is responsi- ble for several interesting features, such as the negative sol- vatochromism and the solution luminescence, while it is as- sociated with large, second-order molecular hyperpolariz- abilities. This state arises from the transition between a metal/CO localized HOMO–2 and a diimine LUMO (both of b 1 symmetry in the case of the C 2v point group). [4,5] Thus the electronic properties of M(CO) 4 (diimine) make their use quite essential in areas such as the investigation of polyme- rization processes, [2] the labeling of biomolecules, [6] and the exploitation of solar energy. [7] All the structures of tetracarbonyl diimine compounds, which are either crystallographically elucidated or theoretic- ally predicted, share two common features. Both of these [a] Inorganic Chemistry Laboratory, Department of Chemistry, National and Kapodistrian University of Athens, Panepistimiopolis, Zografou 15771, Greece Fax: +30-210-8322828 E-mail: cmitsop@chem.uoa.gr Supporting information for this article is available on the WWW under http://www.eurjic.org or from the author. © 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Eur. J. Inorg. Chem. 2007, 110–119 110 proposed that the correct indices of the extent of back-do- nation are the deviations of the C–W–C angles from orthogo- nality and the O–C–W angles from linearity. It is indicated for the first time that the larger the existing deviation the stronger the back-donation. Moreover, a structure-to-proper- ties relation is provided that correlates the bending of the carbonyls to the extent of back-donation. Back-donation is expected to be correlated to the electronic properties of the compounds, such as the extent of solvatochromism. (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2007) features are related to the nonorthogonality of the W(CO) 4 fragment. More specifically, according to all pub- lished structures, the axial carbonyls bend away from the diimine moiety. Moreover, in the majority of the known structures, the two equatorial carbonyl groups bend away from each other. In both cases the linearity of the two CO units is no longer valid. [4–5,8–9] It is also worth mentioning that the M–C–O angle often changes upon electronic exci- tation. [4] These above-mentioned two features have been ob- served several times in the past, but they have never been explained in the literature. Herein, these structural charac- teristics are studied by means of DFT calculations, and our approach is divided into two parts. Firstly, by employing the highly symmetric W(CO) 4 (phen) complex (phen = 1,10- phenanthroline) we attempted to investigate and discuss their structure from an electronic point of view in terms of the π* back-bonding theory. To proceed with our analysis, we performed a fragments’ molecular orbital analysis (FMO) using DFT electron density calculations. Recently, this procedure was successfully used in the case of a related tungsten compound. [5] Secondly, we discussed the same characteristics as they evolved through a series of bidentate N-donor ligands. In the latter case, we succeeded in ex- tracting the structure-to-properties relation (SPR) that gov- erns these compounds. 2. Results and Discussion 2.1. Evolution of Back-Donation in W(CO) 4 (phen) We started our analysis with W(CO) 4 (phen) (1). Despite this complex being one of the most extensively studied