Two new heterometallic copper(II)-mercury(II) complexes [(CuL)Hg(N-3)(2)] (1) and [(CuL)(2)Hg(N-3)(2)] (2) and one copper(II)-cadmium(II) complex [(CuL)(2)Cd(N-3)(2)] (3) have been synthesized using "metalloligand" [CuL] (where H2L = N,N'-bis(salicylidene)-1,3-propanediamine) and structurally characterized. Complex 1 is a one-dimensional (ID) helical coordination polymer constructed by the joining of the dinuclear [(CuL)Hg(N-3)(2)] units through a single mu(-1,1) azido bridge. In the dinuclear unit the Hg(II) is bonded with two phenoxido oxygen atoms of "metalloligand" [CuL] and two nitrogen atoms of azido ligands. Complex 2 is a linear trinuclear entity, in which two terminal "metalloligands" [CuL] are coordinated to central Hg(II) through double phenoxido bridges. The azido ligands link the central mercury atom with the terminal copper atoms via mu(-1,3) bridges. In contrast, the trinuclear complex 3 is bent Here, in addition to two double phenoxido bridges, central Cd(II) is bonded to two mutually cis nitrogen atoms of two terminal azido ligands. The variation in the coordination modes of the azido ligand seems to be responsible for the different molecular shapes of 2 and 3. Interestingly, bond distances between the Hg atoms and the central nitrogen atom of the azido ligands are 2.790(4) and 2.816(5) angstrom in 1 and 2.823(4) angstrom in 2. These bond distances are significantly less than the sum of van der Waals radii of mercury (2.04 angstrom) and nitrogen (1.55 angstrom) and considerably longer than the sum of their covalent radii (2.03 angstrom). However the distances are similar to reported Hg-N bond distances of some Hg(II) complexes. Therefore, we have performed a theoretical density functional theory study to know whether there is any interaction between the central nitrogen atom of the azido ligand and the mercury atoms. We have used the Bader's "atoms-in-molecules", energetic and orbital analyses to conclude that such interaction does not exist. The probable reason for different molecular shapes observed in trinuclear complexes of 2 and 3 also has been studied and explained by theoretical calculations and using the CSD. Electronic spectra, EPR spectra and ESI mass spectra show that all three complexes lose their solid state identity in solution.