Graham and Sloane minimize the second moment of the centres of a number discs in order to maximize their compactness. They use computational geometry techniques to find the optimal packings for non-trivial $n$.
I have adopted this metric in my research. One of the reviewers of my paper insists that there must be a closed-form solution for the lower-bound of this second moment (i.e. the optimal packing) for general $n$, because the problem "seems simple".
I do not believe that this is so. Can you please confirm (or prove me wrong)? A reference confirming the unsolved status of this problem would be most welcome.