“…This approach previously used for a variety of systems [28,[34][35][36][37][38] does not fit our experimental data At T = 300 K, the Fe 15 Ag 85 (see top panel Figure 5) signal is very close to the structural one, being different in the region 0.015 ≤ Q ≤ 0.2Å −1 (highlighted in the top panel of Figure 6), which corresponds to the lengthscale 3.1 ≤ D ≤ 42 nm. This implies some short-range correlations among clusters of Fe nanoparticles (approximately between 2 and 15 closed packed Fe nanoparticles, bearing in mind that the size of a single MNP is around 2.7 nm), even in the most dilute case, at high temperatures and in absence of field, as was suggested in previous studies [4,8,10]. Cooling down to 150 K produces an increase of the magnetic signal in the whole Q-range, specially in the 0.01 ≤ Q ≤ 0.2Å −1 (3.1 ≤ D ≤ 63 nm, clusters spanning 2 to 23 Fe nanoparticles) region.…”