Dear, It's not a metallic system.. it is a crystal of organic molecules. Like in the article there are four amino acids in the unit cell.
Could it be the eggbox effect? I'm performing a new optimization with higher k-points, and the result is the same.. the lattice parameters increase during ~650 CG steps. After it, the parameters decrease outcell: 14.051222 7.857754 5.223328outcell: 14.052645 7.858561 5.223709outcell: 14.054069 7.859367 5.224090outcell: 14.055492 7.860173 5.224471outcell: 14.056915 7.860979 5.224852outcell: 14.058338 7.861786 5.225233outcell: 14.057030 7.861045 5.224883outcell: 14.056670 7.861153 5.225075outcell: 14.056310 7.861261 5.225267outcell: 14.055949 7.861369 5.225459outcell: 14.055589 7.861478 5.225650
-- SIESTA is supported by the Spanish Research Agency (AEI) and by the European H2020 MaX Centre of Excellence (http://www.max-centre.eu/)