Treatment of MCl2 (M = Mn, Fe, Co) with 4 equivalents of LiN=CtBu2 affords the four coordinate tetrahedral `ate' complexes [Li(THF)]2[M(N=CtBu2)4] in high yields. These complexes can be oxidized by 0.5 equiv of I2 to generate the M(III) `ate' complexes [Li(12-crown-4)2][M(N=CtBu2)4] in high yields. These complexes exhibit high spin, intermediate spin, and low spin magnetizations for Mn, Fe, and Co respectively. [Li(THF)]2[M(N=CtBu2)4] can be oxidized by 1.0 equiv of I2 to generate rare examples of neutral M(IV)(N=CtBu2)4 complexes in moderate to high yields. Thermolysis of M(IV)(N=CtBu2)4 (M = Mn, Fe, Co) complexes show the decomposition pathway to proceed through ligand oxidation and the formation of tert-butyl cyanide and tert-butyl radical. Lower valent metal products can be isolated for Mn and Fe yielding Mn3(N=CtBu2)6 and Fe(II)Fe(III)(N=CtBu2)5 in moderate yields.

The addition of 0.25 P4 to Fe(N=CtBu2)4 followed by heating to 60 °C yields Fe(II)Fe(III)(N=CtBu2)5 and PtBu3 confirming the formation of tert-butyl radical. Treatment of Fe(IV)(N=CtBu2)4 with acetylacetone (acac) fails to afford a heteroleptic Fe(IV) complex, but instead generates Fe(N=CtBu2)2(acac) in moderate yields. Additional protonolysis experiments indicate any disruption to the Fe(IV)(N=CtBu2)4 coordination sphere results in decomposition by ligand oxidation. Treatment of MCl2 (M = Mn, Fe, Co) with 2.5 equivalents of LiN=CtBu2 in the presence of 1 equiv of 12-crown-4 affords the dimeric `ate' complexes [Li(12-crown-4)2][M2(N=CtBu2)5] in high yields. In the solid state, these complexes exhibit short M - M distances indicative of metal-metal bonding interactions. Magnetic studies show strong anti-ferromagnetic coupling between metals for all three complexes consistent with metal-metal bonding interactions.

Treatment of FeCl2 with 3 equivalents of LiN=CPh2 or LiN=Fluorenyl (Fluorenyl = C13H8) in the presence of 0.5 equiv of I2 affords the dimeric complexes Fe2(micro-N=CPh2)2(N=CPh2)4 or Fe2(micro-N=Fluorenyl)2(N=Fluorenyl)4 in high and moderate yields respectively. Though these ligands are very similar, in the solid state, the Fe - Fe distance in the N=Fluorenyl complex is 0.4 A shorter and indicative of an Fe(III)-Fe(III) metal-metal bond. The proposed Fe-Fe bonding is supported by solid state magnetometry.