Structure and Bonding in Magnesium Difluoride Clusters: The (MgF2)n(n= 2−3) Clusters

Significant bonding rearrangements triggered by Mg4 clusters

The structure, stability, and bonding of the complexes formed by the interaction of Mg₄ clusters and first row Lewis bases, namely, ammonia, water, and hydrogen fluoride, have been investigated through the use of high-level G4 single-reference and CASPT2 multireference formalisms. The adducts formed reflect the high electrophilicity of the Mg₄ cluster through electron density holes in the neighborhood of each metallic center. After the adduct formation, the metallic bonding of the Mg₄ moiety is not significantly altered so that the hydrogen shifts from the Lewis base toward the Mg atoms lead to new local minima with enhanced stability. For the particular case of ammonia and water, the global minima obtained when all the hydrogens of the Lewis base are shifted to the Mg₄ moiety have in common a very stable scaffold with a N or an O center covalently tetracoordinated to the four Mg atoms, so the initial bonding arrangements of both reactants have completely disappeared. The reactivity features exhibited by these Mg₄ clusters suggest that nanostructures of this metal might have an interesting catalytic behavior.

Global optimization of ionic Mg(n)F(2n) (n=1-30) clusters

The global optimization basin-hopping (BH) method has been used to locate the global minima (GM) of Mg(n)F(2n) (n=1-30) clusters using a Born-Mayer-type potential. Some of the GM were particularly difficult to find, requiring more than 1.5 x 10(4) BH steps. We have found that both the binding energy per MgF2 unit and the effective volume of the GM isomers increase almost linearly with n, and that cluster symmetry decreases with cluster size. The data derived from the BH runs reveal a growing density of local minima just above the GM as n increases. Despite this, the attraction basin around each GM is relatively large, since after all their atomic coordinates are randomly displaced by values as high as 2.0 bohrs, the perturbed structures, upon reoptimization, relax back to the GM in more than 50% of the cases (except for n=10 and 11). The relative stabilities derived from energy second differences suggest that n=8,10,13,15, and 20 are probably the magic numbers for these systems. Mass spectrum experiments would be very useful to clarify this issue.