Metal enolates are an important class of reactive intermediates widely employed in organic synthesis.[1] In contrast, little is known about silenolates, the silicon analogues of enolates.[2, 3] Enolates exist in two tautomeric forms, the enol form and the keto form, and their reactions reflect the coexistence of these two forms.[1] The dominant structure of alkali metal enolates is the enol form both in nonsolvating media and in various solvating media such as THF, N, N, N’, N’-tetramethylethylenediamine, and [18] crown-6.[1] Silenolates also exist in two tautomeric forms: the keto form a (acyl silyl anion) and the enol form b (silene)[Eq.(1)], and they also show ambident reactivity.[2] The first silenolate (solvated), recently isolated and characterized by X-ray crystallography, has the keto form a.[3] An enol-form silenolate b was not yet reported. Isolation of an enol-form silenolate is challenging, because it has a Si= C π bond which is thermodynamically and kinetically less stable than a C= C bond.[4, 5] In addition, enol-form silenolates can be regarded as functional silenes, which are reagents of growing importance in silicon chemistry.[6] Here we report the synthesis, isolation, and X-ray molecular structure of the first enol-form silenolates (tBu-Me2Si) 2Si= C (OLi) Ad (1) and (tBu2MeSi) 2Si= C (OLi) Ad (2). We show by DFT quantum-mechanical calculations that, in contrast to organic enolates, which exist predominantly in the enol form regardless of solvation,[1] the structure of silenolates 1 and 2 is strongly dependent on the solvent. Silenolate 1 was synthesized by metal–halogen exchange between tBuMe2SiLi (in excess) and bromo acyl silane Br (tBuMe2Si) 2SiC (O) Ad (3) in hexane at À788C. Upon warming to room temperature pale yellow crystals of silenolate 1 precipitated (10% yield).[7] The major product is substitution product 4 [Eq.(2)].[7]

The molecular structure of 1 was determined by X-ray crystallography.[8] In the crystal, 1 is an aggregate of three (tBuMe2Si) 2Si= C (OLi) Ad molecules forming a six-membered ring, in which each Li atom is coordinated to two oxygen atoms, and each oxygen atom to two Li atoms (Figure 1). The Si= C bond length in 1 of 1.822 is longer than the Si= C bond (1.764) in (Me3Si) 2Si= C (OSiMe3) Ad (5),[9] probably because of the larger substituents in 1 compared to 5,[10] but it is significantly shorter than the SiÀC bond length (1.926) in acyl silyl anion[(Me3Si) 2SiC (O) tBu] À K+[18] crown-6 (6).[3] The essentially planar Si1 atom of 1 (Σq (Si1)= 359.258) suggests sp2 hybridization. The Si3-Si1-C1-O1 torsion angle in 1 of 9.38 is smaller than the analogous torsion angle in 5 (14.68).[9] The OÀLi bond lengths (1.858 and 1.834) in 1 are in the regular range of OÀLi bond lengths in enolates (1.80–1.90).[11] Thus, 1 has the expected structure of an enol-form silenolate b, with an Si= C bond. Aiming to obtain an enol form silenolate in higher yield, we prepared bromo acyl silane 3a, an analogue of 3 with larger substituents (tBu2MeSi in 3a vs. tBuMe2Si in 3). Reaction of 3a with tBuMe2SiLi at À788C in hexane followed by warming to room temperature gave silenolate 2 as a yellow crystalline powder in 80% yield [Eq.(3)].[7b] The structure of 2 was determined by X-ray crystallography (Figure 2).[12] Compound 2 is a dimer of a coaggregate of silenolate with tBuMe2SiLi, that is,[(tBu2MeSi) 2Si= C (OLi) Ad· tBuMe2SiLi] 2. The structure of the silenolate part of 2 resembles the structure of 1 (Figure2b), although it has a different