可转移的深度学习潜力揭示了LiF-NaF-ZrF4熔盐中的中程有序效应
Transferable Deep Learning Potential Reveals Intermediate-Range Ordering Effects in LiF-NaF-ZrF4 Molten Salt
摘要:
LiF-NaF-ZrF 4多元熔盐具有良好的热物理和输运性能,是先进清洁能源系统的理想冷却剂。然而,复杂的结构,使这些性能,以及它们对组合物的依赖,几乎是量化的,由于在模拟和解释实验光谱的高度无序,中程结构的限制。具体而言,尺寸有限的从头计算模拟和精度有限的经典模型在过去使用,是无法捕捉到广泛的波动图案中发现的扩展的异质结构的液体盐。这极大地抑制了我们设计定制组合物和材料的能力。在这里,准确,高效和可转移的机器学习潜力用于预测LiF-NaF-ZrF 4中远远超出第一配位壳层的结构。神经网络训练只在低共熔组合物与29%和37%的ZrF 4被证明准确地模拟了广泛的组合物(11至40%的ZrF 4)具有显着不同的配位化学,同时显示出显着的协议与理论和实验拉曼光谱。理论拉曼计算进一步揭示了在~250 cm-1处的弯曲带的先前不可见的移位和平坦化,这验证了在具有高于29%ZrF4含量的组合物中观察到的模拟的扩展范围结构。在这种情况下,基于机器学习的模拟能够访问更大的时间和长度尺度(超过17 μ m),对于准确预测结构和离子扩散率至关重要。
英文摘要:
LiF-NaF-ZrF4 multicomponent molten salts are promising candidate coolants for advanced clean energy systems owing to their desirable thermophysical and transport properties. However, the complex structures enabling these properties, and their dependence on composition, is scarcely quantified due to limitations in simulating and interpreting experimental spectra of highly disordered, intermediate-ranged structures. Specifically, size-limited ab-initio simulation and accuracy-limited classical models used in the past, are unable to capture a wide range of fluctuating motifs found in the extended heterogeneous structures of liquid salt. This greatly inhibits our ability to design tailored compositions and materials. Here, accurate, efficient and transferable machine learning potentials are used to predict structures far beyond the first coordination shell in LiF-NaF-ZrF4. Neural networks trained at only eutectic compositions with 29% and 37% ZrF4 are shown to accurately simulate a wide range of compositions (11 to 40% ZrF4) with dramatically different coordination chemistries, while showing a remarkable agreement with theoretical and experimental Raman spectra. The theoretical Raman calculations further uncovered the previously unseen shift and flattening of bending band at ~250 cm-1 which validated the simulated extended-range structures as observed in compositions with higher than 29% ZrF4 content. In such cases, machine learning-based simulations capable of accessing larger time- and length-scales (beyond 17 Å) were critical for accurately predicting both structure and ionic diffusivities.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。