目前的材料科學家一般透過分析一系列顯微照片來研究或描述工程材料的特性，包括從毫米到奈米的複雜微觀結構。這些工作通常是由科學家個人手動完成的，有時還需要計算技術的輔助。這些以人為中心的工作流程存在嚴重的缺點，如對專業要求高、可重複性差、過程耗時長等。以奈米級L12型有序結構為例，該結構被廣泛用於面心立方（FCC）合金中，以利用其硬化能力，從而提高機械效能。這些細尺度的顆粒通常與具有相同原子構型、不考慮化學種類的基體完全相干，這使得他們的表徵具有挑戰性。空間分佈圖（SDMs）用於透過詢問重建原子探針斷層掃描（APT）資料內原子的三維（3D）分佈來探究區域性秩序。然而，手動分析完整的點雲（> 1000萬個）以尋找資料中保留的部分晶體學資訊，幾乎是不可能的。

來自德國馬普所的Yue Li和Leigh T. Stephenson等提出了一種基於卷積神經網路(CNNs)的策略，利用APT資料自動識別FCC基合金中的奈米級L12型有序結構，具有超高的識別能力。該方法首先生成了模擬L12有序結構的SDMs和FCC矩陣。這些模擬影象結合少量的實驗資料，用於訓練基於CNN的L12有序結構識別模型。最後，成功應用該方法揭示了FCC Al-Li-Mg體系中平均半徑為2.54 nm的L12型δ'-Al3（LiMg）奈米顆粒的3D分佈。可檢測得奈米域最小半徑甚至低至5 Å。所提出的CNN-APT方法很有希望在不久的將來擴充套件到識別其他奈米級的有序結構，甚至更有挑戰性的短程有序現象中。

該文近期發表於npj Computational Materials 7: 8 (2021)，英文標題與摘要如下，點選https://www.nature.com/articles/s41524-020-00472-7可以自由獲取論文PDF。

Convolutional neural network-assisted recognition of nanoscale L12 ordered structures in face-centred cubic alloys

Yue Li, Xuyang Zhou, Timoteo Colnaghi, Ye Wei, Andreas Marek, Hongxiang Li, Stefan Bauer, Markus Rampp & Leigh T. Stephenson

Nanoscale L12-type ordered structures are widely used in face-centered cubic (FCC) alloys to exploit their hardening capacity and thereby improve mechanical properties. These fine-scale particles are typically fully coherent with matrix with the same atomic configuration disregarding chemical species, which makes them challenging to be characterized. Spatial distribution maps (SDMs) are used to probe local order by interrogating the three-dimensional (3D) distribution of atoms within reconstructed atom probe tomography (APT) data. However, it is almost impossible to manually analyze the complete point cloud (>10 million) in search for the partial crystallographic information retained within the data. Here, we proposed an intelligent L12-ordered structure recognition method based on convolutional neural networks (CNNs). The SDMs of a simulated L12-ordered structure and the FCC matrix were firstly generated. These simulated images combined with a small amount of experimental data were used to train a CNN-based L12-ordered structure recognition model. Finally, the approach was successfully applied to reveal the 3D distribution of L12–type δ′–Al3(LiMg) nanoparticles with an average radius of 2.54 nm in a FCC Al-Li-Mg system. The minimum radius of detectable nanodomain is even down to 5 Å. The proposed CNN-APT method is promising to be extended to recognize other nanoscale ordered structures and even more-challenging short-range ordered phenomena in the near future.