In this work we propose to use a continuous Variable Neighborhood Search (VNS for short) heuristic for minimizing the potential energy function of molecules. Computing the global minimum of this function is very difficult because it has a large number of local minima which grow exponentially with molecule size. Experimental evidence shows that in the great majority of cases the global minimum potential energy of a given molecule corresponds to its tridimensional structure and this structure is important because it dictates most of the properties of the molecule. Computational results for problems with up to 200 degrees of freedom are presented.