Low coordinate, monomeric molybdenum and tungsten(III) complexes: structure, reactivity and calculational studies of (silox)3Mo and (silox)3ML (M = Mo, W; L = PMe3, CO; silox = (t)Bu3SiO).

Treatment of (silox)3MCl (M = Mo, 1-Cl; W, 2-Cl; silox = (t)Bu3SiO) with PMe3 and Na/Hg led to formation of monomeric, d(3) phosphine adducts, (silox)3MPMe3 (M = Mo, 1-PMe3; W, 2-PMe3) via (silox)3ClMPMe3 (M = Mo, 1-ClPMe3; W, 2-ClPMe3). Structural studies show 1-PMe3 and 2-PMe3 to be highly distorted; calculations on full chemical models corroborate experimentally determined S = 1/2 ground states and their structural features. The compounds contain a bent M-P bond that is characteristic of significant sigma/pi-mixing. PMe3 may be thermally removed from 1-PMe3 in vacuo to produce (4)A2' (silox) 3Mo (1), which was derivatized with CO, NO, and 1/4 P4 to form (silox)3Mo (1-CO), (silox)3MoNO (1-NO), and (silox)3MoP (1-P), respectively. Calculations revealed (silox)3W (2') to have an S = 1/2 ground state, which may render it too reactive to be isolated. Treatment of 2-PMe3 with CO, NO, and 1/4 P4 formed (silox)3WCO (2-CO), (silox)3WNO (2-NO), and (silox)3WP (2-P), respectively. 2-CO and 2-NO are more conveniently prepared from Na/Hg reductions of 2-Cl in the presence of CO and NO, respectively. Calculations reveal subtle effects of nd(z2)/(n+1)s mixing in differentiating the chemistry of Mo and W and in rationalizing the generation of mononuclear species.