有機-無機鈣鈦礦材料，特別是甲基胺鹵化鉛(MAPbX3, X=I, Br, Cl)，因其良好的光電性質而在太陽能電池和光電探測器等領域受到廣泛關注。然而，這類材料的獨特的性質與其複雜的微觀動力學密切相關，而這些原子層面的機制尚未被徹底理解。第一原理分子動力學(AIMD)受計算資源限制，難以模擬足夠長的時間尺度和大規模的系統來呈現這些動力學行為，而經驗勢能往往缺乏足夠的精度和泛化能力。因此，開發精確高效的原子模擬工具對於深入理解這類材料的微觀物理機制及促進其應用發展相當重要。
本研究建立能精確描述MAPbBr3的機器學習勢能模型，我們採用了深度神經網路的Deep Potential (DP)模型，DP模型能夠在保持對稱性的前提下，精確描述原子間複雜的多體相互作用。為解決訓練數據生成的問題，我們利用了自動化的Deep Potential GENerator (DP-GEN)流程，DP-GEN透過不斷的迭代循環，將高精度的第一原理計算與模型篩選結合，高效地探索構型空間並生成訓練數據集，所訓練出的DP模型與第一原理計算數據的詳盡比較及對物理性質的驗證，證實其足夠準確。
利用此成功驗證的機器學習勢能模型，我們進行大規模、長時間尺度的分子動力學模擬，成功突破AIMD的限制，這些模擬使我們能夠深入研究MAPbBr3在不同溫度下的微觀動力學行為，並詳細分析了MA+陽離子的取向分布、轉動弛豫時間等關鍵動力學參數，研究結果清楚展示了MA+陽離子轉動行為隨溫度而變化，尤其在低溫下受限制的轉動與高溫下的隨機取向之間存在顯著的轉變，其行為與實驗觀察高度吻合，從而更進一步理解其中微觀動力學過程，並有助於推動有機-無機鈣鈦礦材料的發展與創新。

Organic-inorganic halide perovskites, such as methylammonium lead halides (MAPbX3, X=I, Br, Cl), are highly promising for optoelectronic applications due to their excellent properties. However, a complete understanding of their complex microscopic dynamics, which underpin these unique characteristics, remains elusive. Ab-Initio Molecular Dynamics (AIMD) is often computationally prohibitive for the necessary timescales and system sizes, while empirical potentials typically lack sufficient accuracy and generalizability.
To address this, a machine learning potential model for MAPbBr3 was developed using a deep neural network-based Deep Potential (DP) model. This DP model accurately captures complex many-body interactions while preserving symmetry. An automated Deep Potential Generator (DP-GEN) workflow efficiently generated diverse, high-quality training datasets by iteratively combining high-accuracy first-principles calculations with model selection. The trained model was rigorously validated against first-principles data and various physical properties.
Leveraging this validated model, large-scale, long-timescale molecular dynamics simulations were conducted, extending beyond AIMD limitations. These simulations enabled an in-depth investigation of microscopic dynamic behavior in MAPbBr3 across various temperatures. Detailed analysis of MA+ cation orientation distribution and rotational relaxation times revealed a clear temperature-dependent evolution, transitioning from restricted rotation at low temperatures to random orientation at high temperatures. This observed behavior highly aligns with experimental observations, contributing to a more profound understanding of the material's underlying microscopic dynamic processes.

1. Faghihnasiri, M., M. Izadifard, and M.E. Ghazi, DFT study of mechanical properties and stability of cubic methylammonium lead halide perovskites (CH3NH3PbX3, X= I, Br, Cl). The Journal of Physical Chemistry C, 2017. 121(48): p. 27059-27070.
2. Laamari, M.E., et al., Optimized opto-electronic and mechanical properties of orthorhombic methylamunium lead halides (MAPbX3)(X= I, Br and Cl) for photovoltaic applications. Solar Energy, 2019. 182: p. 9-15.
3. Wang, H., L. Zhang, and J. Han, DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics. Computer Physics Communications, 2018. 228: p. 178-184.
4. Maity, S., et al., Deciphering the nature of temperature-induced phases of MAPbBr3 by ab initio molecular dynamics. Chemistry of Materials, 2022. 34(23): p. 10459-10469.
5. Hata, T., et al., Development of a classical interatomic potential for MAPbBr3. The Journal of Physical Chemistry C, 2017. 121(7): p. 3724-3733.
6. Mattoni, A., et al., Methylammonium rotational dynamics in lead halide perovskite by classical molecular dynamics: the role of temperature. The Journal of Physical Chemistry C, 2015. 119(30): p. 17421-17428.
7. Gordon, R., On the rotational diffusion of molecules. The Journal of Chemical Physics, 1966. 44(5): p. 1830-1836.
8. Zhu, H., et al., Screening in crystalline liquids protects energetic carriers in hybrid perovskites. Science, 2016. 353(6306): p. 1409-1413.
9. Wasylishen, R.E., O. Knop, and J.B. Macdonald, Cation rotation in methylammonium lead halides. Solid state communications, 1985. 56(7): p. 581-582.
10. Poglitsch, A. and D. Weber, Dynamic disorder in methylammoniumtrihalogenoplumbates (II) observed by millimeter-wave spectroscopy. Journal of Chemical Physics, 1987. 87(11): p. 6373-6378.
11. De Graef, M. and M.E. McHenry, Structure of materials: an introduction to crystallography, diffraction and symmetry. 2012: Cambridge University Press.
12. Bhalla, A.S., R. Guo, and R. Roy, The perovskite structure—a review of its role in ceramic science and technology. Materials research innovations, 2000. 4(1): p. 3-26.
13. Green, M.A., A. Ho-Baillie, and H.J. Snaith, The emergence of perovskite solar cells. Nature photonics, 2014. 8(7): p. 506-514.
14. Shi, L., et al., Periodically ordered nanoporous perovskite photoelectrode for efficient photoelectrochemical water splitting. ACS nano, 2018. 12(6): p. 6335-6342.
15. Li, F., et al., Piezoelectric activity in perovskite ferroelectric crystals. IEEE transactions on ultrasonics, ferroelectrics, and frequency control, 2015. 62(1): p. 18-32.
16. Nuraje, N. and K. Su, Perovskite ferroelectric nanomaterials. Nanoscale, 2013. 5(19): p. 8752-8780.
17. Wu, T. and P. Gao, Development of perovskite-type materials for thermoelectric application. Materials, 2018. 11(6): p. 999.
18. Gu, C. and J.-S. Lee, Flexible hybrid organic–inorganic perovskite memory. ACS nano, 2016. 10(5): p. 5413-5418.
19. Younis, A., et al., Halide perovskites: a new era of solution‐processed electronics. Advanced Materials, 2021. 33(23): p. 2005000.
20. Herz, L.M., Charge-carrier dynamics in organic-inorganic metal halide perovskites. Annual review of physical chemistry, 2016. 67(1): p. 65-89.
21. Leguy, A.M., et al., Dynamic disorder, phonon lifetimes, and the assignment of modes to the vibrational spectra of methylammonium lead halide perovskites. Physical Chemistry Chemical Physics, 2016. 18(39): p. 27051-27066.
22. Dai, T., et al., Strategies for high-performance large-area perovskite solar cells toward commercialization. Crystals, 2021. 11(3): p. 295.
23. Veldhuis, S.A., et al., Perovskite materials for light‐emitting diodes and lasers. Advanced materials, 2016. 28(32): p. 6804-6834.
24. Kim, Y.H., et al., Multicolored organic/inorganic hybrid perovskite light‐emitting diodes. Advanced materials, 2015. 27(7): p. 1248-1254.
25. Ahmadi, M., T. Wu, and B. Hu, A review on organic–inorganic halide perovskite photodetectors: device engineering and fundamental physics. Advanced Materials, 2017. 29(41): p. 1605242.
26. Pisoni, A., et al., Ultra-low thermal conductivity in organic–inorganic hybrid perovskite CH3NH3PbI3. The journal of physical chemistry letters, 2014. 5(14): p. 2488-2492.
27. Jin, H., et al., Hybrid organic–inorganic thermoelectric materials and devices. Angewandte Chemie International Edition, 2019. 58(43): p. 15206-15226.
28. Vijayakanth, T., et al., Recent advances in organic and organic–inorganic hybrid materials for piezoelectric mechanical energy harvesting. Advanced Functional Materials, 2022. 32(17): p. 2109492.
29. McGovern, L., et al., Understanding the stability of MAPbBr3 versus MAPbI3: suppression of methylammonium migration and reduction of halide migration. The journal of physical chemistry letters, 2020. 11(17): p. 7127-7132.
30. Lee, J.-W., et al., Dynamic structural property of organic-inorganic metal halide perovskite. Iscience, 2021. 24(1).
31. Egger, D.A., A.M. Rappe, and L. Kronik, Hybrid organic–inorganic perovskites on the move. Accounts of chemical research, 2016. 49(3): p. 573-581.
32. Liu, S., R. Guo, and F. Xie, The effects of organic cation rotation in hybrid Organic-Inorganic Perovskites: A critical review. Materials & Design, 2022. 221: p. 110951.
33. Lin, C.-C., et al., Direct investigation of the reorientational dynamics of A-site cations in 2D organic-inorganic hybrid perovskite by solid-state NMR. Nature Communications, 2022. 13(1): p. 1513.
34. Wang, K.-H., et al., Structural and photophysical properties of methylammonium lead tribromide (MAPbBr3) single crystals. Scientific reports, 2017. 7(1): p. 13643.
35. Mannino, G., et al., Temperature-dependent optical band gap in CsPbBr3, MAPbBr3, and FAPbBr3 single crystals. The journal of physical chemistry letters, 2020. 11(7): p. 2490-2496.
36. Ahmadi, M., et al., Spatially resolved carrier dynamics at MAPbBr3 single crystal–electrode interface. ACS applied materials & interfaces, 2019. 11(44): p. 41551-41560.
37. Fu, J., et al., Localized traps limited recombination in lead bromide perovskites. Advanced Energy Materials, 2019. 9(12): p. 1803119.
38. Wang, C., et al., Environmental surface stability of the MAPbBr3 single crystal. The Journal of Physical Chemistry C, 2018. 122(6): p. 3513-3522.
39. Wei, K., et al., Temperature-dependent excitonic photoluminescence excited by two-photon absorption in perovskite CsPbBr3 quantum dots. Optics letters, 2016. 41(16): p. 3821-3824.
40. Tilchin, J., et al., Hydrogen-like Wannier–Mott excitons in single crystal of methylammonium lead bromide perovskite. ACS nano, 2016. 10(6): p. 6363-6371.
41. Hohenberg, P. and W. Kohn, Inhomogeneous electron gas. Physical review, 1964. 136(3B): p. B864.
42. Kohn, W. and L.J. Sham, Self-consistent equations including exchange and correlation effects. Physical review, 1965. 140(4A): p. A1133.
43. Zhang, L., et al., End-to-end symmetry preserving inter-atomic potential energy model for finite and extended systems. Advances in neural information processing systems, 2018. 31.
44. Behler, J., Atom-centered symmetry functions for constructing high-dimensional neural network potentials. The Journal of chemical physics, 2011. 134(7).
45. Bartók, A.P., R. Kondor, and G. Csányi, On representing chemical environments. Physical Review B—Condensed Matter and Materials Physics, 2013. 87(18): p. 184115.
46. Bartók, A.P., et al., Gaussian approximation potentials: The accuracy of quantum mechanics, without the electrons. Physical review letters, 2010. 104(13): p. 136403.
47. Weyl, H., The classical groups: their invariants and representations. Vol. 1. 1946: Princeton university press.
48. Zaheer, M., et al., Deep sets. Advances in neural information processing systems, 2017. 30.
49. Kingma, D.P. and J. Ba, Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
50. Behler, J. and M. Parrinello, Generalized neural-network representation of high-dimensional potential-energy surfaces. Physical review letters, 2007. 98(14): p. 146401.
51. Williams, C.K. and C.E. Rasmussen, Gaussian processes for machine learning. Vol. 2. 2006: MIT press Cambridge, MA.
52. Hofmann, T., B. Schölkopf, and A.J. Smola, Kernel methods in machine learning. 2008.
53. Zhang, Y., et al., DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models. Computer Physics Communications, 2020. 253: p. 107206.
54. Jain, A., et al., Commentary: The Materials Project: A materials genome approach to accelerating materials innovation. APL materials, 2013. 1(1).
55. Thompson, A.P., et al., LAMMPS-a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales. Computer physics communications, 2022. 271: p. 108171.