{N,N-Bis[2-(trimethylsilylamino)ethyl]-N′-(trimethylsilyl)ethane-1,2-diaminato(3–)-κ4 N}methylzirconium(IV)

The title compound, [Zr(CH3)(C15H39N4Si3)], is a unique example of a triamidoamine-supported zirconium–methyl complex that crystallized as a monomer with trigonal–bipyramidal geometry at zirconium, featuring a Zr—C bond of 2.2963 (16) Å.


Comment
Simple derivatives of triamidoamine-supported zirconium have recently been prepared, and these complexes have been found to be effective pre-catalysts for dehydrocoupling involving phosphines (Roering et al., 2007;Waterman, 2007) and arsines (Roering et al., 2008). Reaction of (N 3 N)ZrCl with MeLi in Et 2 O afforded (N 3 N)ZrMe (I, N 3 N = N(CH 2 CH 2 NSiMe 3 ) 3 3-)) in 82% yield as colorless crystals (Waterman, 2007). Complex I is highly lipophilic and exhibits reasonable thermal stability despite a low melting temperature (Mp = 87-91°C). Upon heating in benzene solution, methane is evolved, and the triamidolamine ligand is metalated at the trimethylsilyl functionality-a key step in dehydrocoupling catalysis (Waterman, 2007). Whereas zirconium complexes with methyl ligands are commonplace, those supported by triamidoamine ligands were previously unknown.
Complex I was crystallized from cold pentane solution and exhibits distorted trigonal bipyramidal geometry at zirconium.
The τ5-parameter for the complex was calculated as 1.1 (Addison et al., 1984), though greater than 1, it is consistent with a trigonal bipyramidal geometry rather than square-pyramidal. The value greater than unity is attributed to the fact that the amido nitrogen atoms are not co-planar with Zr as expected for a trigonal pyramid, a feature that artificially increases the α angle in the τ5-parameter calculation.
For the methide ligand, the Zr-C bond length (2.2963 (16) Å) of I is slightly shorter than the most closely related complex structurally characterized complex (N 3 N*)ZrCH 2 Ph (N 3 N* = N(CH 2 CH 2 NSiMe 2 t Bu) 3 3-), which displays Zr-C 2.3243 (18) Å (Morton et al., 1999). Of the ca 315 complexes featuring a Zr-CH 3 bond, the Zr-C bond of I fits neatly into the range of reported Zr-C bond lengths of 2.046 to 2.805 Å.
The Hirshfeld Test Difference for Zr-C(16) is 6.71 su. This value is most likely the result of a small amount of contamination by the (N 3 N)ZrCl precursor. This is an interesting observation given the high analytical purity of I and the inability to observe any (N 3 N)ZrCl in benzene-d 6 solutions of I by 1 H NMR spectroscopy. However, residual electron density was observed along the Zr-C vector, but this peak could not be refined as a chloride disorder. Contamination of crystals by trace quantities of chloride derivatives is known, and due caution must be taken in evaluating data (Parkin, 1992). While this test does not significantly detract from the quality of this structure and solution and correlation between the structure and spectroscopic assignment, it does indicate that contamination of product complexes with (N 3 N)ZrCl is possible. This supplementary materials sup-2 observation suggests that synthetic strategies for (N 3 N)ZrX derivatives that circumvent the use of (N 3 N)ZrCl are optimal for achieving highly pure compounds for catalytic applications.

Experimental
Methyl complex (N 3 N)ZrMe (I) was prepared according to the literature procedure (Waterman, 2007). Dissolving complex I (ca 800 mg) in minimal pentane (2 ml), then filtering and cooling the resulting colorless solution to −30 °C for extended periods (ca 1-2 weeks) afforded colorless, X-ray quality crystals.

Refinement
Hydrogen atoms on carbon were included in calculated positions and were refined using a riding model [C-H = 0.98 (CH 3 ), 0.99 Å (CH 2 ); U iso = 1.5 U eq (CH 3 ), 1.2 U eq (CH 2 )]. The Hirshfeld test difference value Zr-C(16) = 6.71 su indicates slight contamination with the chloro precursor. The slighly low U eq value for Si (1) is likely the result of the terminal the atom residing in a trimethylsilyl substituent. Crystal data [Zr(CH 3  Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger. EXTI refined to 0 and was removed from the refinement.