Recent collision-induced dissociation studies on the gasphase [Pd 3 (CO) 6 ] À cluster [1] supplement other experimental observations, [2±7] which seem to suggest the feasibility of homoleptic carbonylpalladium derivatives. Thus, although the measured average PdÀCO bond dissociation energy (ca. 134 kJ mol
À1) is weaker than that measured for the [Pt 3 (CO) 6 ] À congener (ca. 173 kJ mol À1 ), its value does not seem, however, sufficient to rule out the possible existence of binary palladium carbonyl compound in solution, unless their formation is hindered by some competing reaction.[1] The belief that the preassembly of a Pd kernel of critical size could ultimately favor the synthesis of a binary palladium carbonyl compound in solution led us to tackle the synthesis of Ni ± Pd homoleptic carbonyl bimetallic clusters (HCBCs) 6 [2] were obtained in low yields (5 ± 10 %) from the filtered THF reaction solution upon layering with n-hexane. The structure of 1 [11] is shown in Figure 1; its metal skeleton is represented in Figure 2 (left) as deriving from condensation of two Ni 8 Pd 8 vertex-truncated n 3 -tetrahedra through their Ni 2 Pd 5 hexagonal faces, giving rise to a six-layer close-packed metal framework. Two Pd atoms are encapsulated in two Ni 2 Pd 10 twinned cubooctahedral moieties and the cluster surface solely comprises triangular faces with all possible combinations of the two metals. The forty carbonyl groups display most possible CO coordination modes. They comprise linear, and symmetrical edge-and face-bridging ligands, as well as unsymmetrical and semibridging carbonyl groups with sub-van-der-Waals PdÀC contacts of 2.3 ± 2.5 . Only nickel atoms bind terminal carbonyl groups. Both homo-and heterometallic edges are spanned by bridging CO groups. Notably, only the Pd 3 triangles of the miscellaneous Ni 3Àx Pd x