多相多孔材料的幾何特性對於很多工程學科來說至關重要,例如,貴金屬催化劑在多孔基底上的分散,高性能合金中金屬相結構和缺陷。在電化學中,無論電池、燃料電池或者超級電容器,它們的電極結構通常是典型的多孔狀,以實現最大化表面積和提供傳輸電子和離子的路徑,與此同時來保持足夠的機械完整度。因此,這些電極材料的微結構和形貌優化對於發展下一代能源存儲技術非常重要。但是,如何提取微觀結構資料庫中關鍵因素或者本質因素,以及如何在不犧牲真實材料總體相似度的前提下調節微觀結構數據的特定屬性仍然存在很多挑戰。
來自英國倫敦帝國理工學院的Cooper教授領導的研究團隊,利用基於生成對抗網絡(GAN)的方法生成了多相三維微結構數據,並將其應用於兩種常見的三相電極(鋰離子電池的陰極和固態氧化物燃料電池陽極),產生了被認為微結構「虛擬表示」的可訓練生成器。他們還進行了真實的和生成的微結構性質的可定量比較,並確定了方法的有效性。作者基於GAN的方法展示了周期邊界對於生成周期微觀結構和擴散模擬的影響。這一方法可以擴展到生成任意大小、多相周期微觀結構,將會引起電化學模擬領域更大的研究興趣。
該文近期發表於npj Computational Materials 6: 82 (2020),英文標題與摘要如下,點擊https://www.nature.com/articles/s41524-020-0340-7可以自由獲取論文PDF。
Pores for thought: generative adversarial networks for stochastic reconstruction of 3D multi-phase electrode microstructures with periodic boundaries
Andrea Gayon-Lombardo, Lukas Mosser, Nigel P. Brandon & Samuel J. Cooper
The generation of multiphase porous electrode microstructures is a critical step in the optimisation of electrochemical energystorage devices. This work implements a deep convolutional generative adversarial network (DC-GAN) for generating realistic nphase microstructural data. The same network architecture is successfully applied to two very different three-phase microstructures: A lithium-ion battery cathode and a solid oxide fuel cell anode. A comparison between the real and synthetic data is performed interms of the morphological properties (volume fraction, specific surface area, triple-phase boundary) and transport properties(relative diffusivity), as well as the two-point correlation function. The results show excellent agreement between datasets and theyare also visually indistinguishable. By modifying the input to the generator, we show that it is possible to generate microstructurewith periodic boundaries in all three directions. This has the potential to significantly reduce the simulated volume required to beconsidered 「representative」 and therefore massively reduce the computational cost of the electrochemical simulations necessary topredict the performance of a particular microstructure during optimisation.