Fig. 4: A reconstitution process to determine the true SRO.
From: Quantifying short-range order using atom probe tomography
a, Measured SRO values (red) versus true SRO values (Supplementary Table 2) embedded in the simulated model of a CoCrNi MEA. This enables the determination of a correction factor β, which accounts for the combined effects of detection loss (57% detection rate) and limited spatial resolution (\({\sigma }_{x,y-}\) = 0.25 nm, \({\sigma }_{z-}\) = 0.10 nm) for each value of SRO for this alloy system (green). The low-SRO range shown in the inset and Fig. 2g was used to determine the go/no-go threshold value for SRO below which random and non-random events are indistinguishable (that is, α < 0.0048). b–d, Validation of the reconstitution procedure via a comparison between (arbitrary) true SRO values (black) to the reconstituted SRO values (blue). Three sets of arbitrary true values were assessed to test high (b), medium (c) and low (d) SRO values (the values are listed in Supplementary Table 3). The data points (blue) show the 95% confidence intervals. 57% of the data was simulated 100 times using the random labelling method with ~4 million atoms, and the SRO was measured for kNN = 7 to generate the range labelled random (violet). Data are presented as the average of the reconstituted SRO value for each pair ± their 95% confidence region. The SRO values for random simulations ranged for |α| ≤ 0.00022.
