“…It is widely believed that as the value of ℓ is increased with all other parameters fixed, the global energy minimizer of E ε should be either constant or spatially periodic, with period approaching a constant independent of ℓ as ℓ → ∞. Proving such a crystallization result would be one of the main challenges in the theory of energy-driven pattern formation and is currently out of reach (for a recent review, see [1]), except for the case d = 1,ū ε ∈ (−1, 1) fixed and ε > 0 sufficiently small [2,22,26] (for a very recent result in that direction in higher dimensions, see [7]). On the other hand, it is known that forū ε ∈ (−1, 1) fixed, global energy minimizers are not constant as soon as ε ≪ 1 and ℓ 1 [3,24].…”