The isolation and characterization of the first molecular complex of divalent thulium, [TmI 2 (dme) 3 ] (dme = 1,2-dimethoxyethane), [1] ten years ago, followed by the dysprosium [2] and neodymium [3] diiodides, [DyI 2 (dme) 3 ] and [NdI 2 (thf) 5 ], opened up a new era in lanthanide chemistry. [4] Hitherto, divalent lanthanide chemistry had been restricted for decades to the "classical" Eu II , Yb II , and Sm II . [4] These "new" divalent complexes have since been investigated in numerous reactions, often revealing unexpected behavior. [5, 6] However, the expectation that these precursors would lead to the isolation of divalent organometallic complexes by salt metathesis reactions was fulfilled only in the case of several very crowded Tm II cyclopentadienyl, phospholyl, and scorpionate complexes. [7] In contrast, all reported attempts with DyI 2 and NdI 2 , or the hypothetical [NdCl 2 (thf) x ], afforded only trivalent products. [7a, 8] An early report on the identification of an organoneodymium(II) complex by elemental analysis, IR spectroscopy, and reactivity [8b] has been questioned. [8c] Lappert and co-workers attempted the synthesis of La II , Ce II , Nd II , and Tm II complexes by reduction of trivalent precursors of the type [Cp R
3 Ln] (Cp R = C 5 H 3 tBu 2 , C 5 H 3 (SiMe 3 ) 2 ) using alkali metals [9] and achieved the isolation of the first crystallographically characterized La II derivatives [9b,e] and recently a mixed Ce II /Ce III complex. [9e] A similar approach used by Cloke and co-workers afforded a mixed Sc I /Sc III compound. [10] The potential of the "new" divalent lanthanide chemistry has recently been highlighted. [11] We extended the reductive approach to obtain stable, solvent-free divalent organothulium complexes carrying cyclopentadienyl or phospholyl ligands by reduction of trivalent precursors of the L 2 LnX type with potassium graphite in nonpolar solvents. [12] This method also gave access to the first divalent organodysprosium ate complexes [Cp ttt 2 Dy(m-X)K([18]crown-6)] (Cp ttt = C 5 H 2 tBu 3 and X = I (1), Br, BH 4 ), which are powerful reducing agents. [13] Despite several attempts at reducing trivalent Nd precursors, no organoneodymium(II) complexes could be isolated, owing to decomposition of any Nd II species in situ by interaction with solvent molecules, such as DME, THF, or benzene. [9a,c-e] We herein report on the synthesis and characterization of the first divalent organoneodymium complex, its behavior towards molecular dinitrogen, and its decomposition by CÀH activation to form a structurally characterized "tuck-in" complex, the first such example for a 4f element. [14] Reduction of the trivalent precursor [Cp ttt 2 NdI] (2), prepared from NdI 3 and [KCp ttt ], with potassium graphite under argon in the presence of [18]crown-6 led to an immediate color change of the solution from green-blue to dark red. After 24 h, dark red crystals of the divalent ate complex [Cp ttt 2 Nd(m-I)K([18]crown-6)]•C 6 H 12 (3), suitable for X-ray crystallography, had formed. (Scheme 1 and Figure 1)
The crystal structure of 3 shows a seven-coordinate neodymium center in an open sandwich array (Cp-Nd-Cp 149.48) with an iodine bridge to the potassium-crown ether complex. The balance of charges indicates that 3 is indeed a Nd II complex that is best described as a [K([18]crown-6)I] adduct of the neutral [Cp ttt
2 Nd] fragment. The geometrical arrangement is similar to the previously reported divalent organodysprosium ate complex 1. [13] The longer detected LnÀI (3.16 for Ln = Nd, 3.06 for Ln = Dy) and LnÀCp(centroid) (2.53 for Ln = Nd, 2.44 for Scheme 1. Synthesis of divalent Nd ate complex 3 and trivalent Nd "tuck-in" complex 4.