Graph representation-based machine learning framework for predicting electronic band structures of quantum-confined nanostructures
Source abstract
Abstract
机器学习技术在电子结构理论中的应用有望实现从头计算精度 的电子性质预测. 然而, 以往的机器学习模型大多只预测本征材料的一 种或几种特定性质. 对包含了电子结构主要信息的能带结构的预测则 还有待深入研究. 因为输入和输出数据的高度可变性, 这是一项颇具挑 战性的任务——输入的材料可以有不同的尺寸和组成成分, 输出的能带 结构可以有数量不等的能带和k点采样. 尤其是考虑量子受限体系时, 因其能带结构对其所受的量子限制十分敏感, 这项任务变得更加困难. 本文提出了一种从量子受限体系的几何结构预测其能带结构的机器学 习框架. 该框架引入的图卷积神经网络可用于具有不同组成成分和几 何形状的材料, 并提取各原子的局域环境信息. 这些信息随后被一个可 学习的实空间哈密顿量构造过程接收并用于预测相关材料在任意k点 的电子结构. 在此构造过程中引入的相关物理理论不仅有助于将量子 限制的微小变化反映到能带结构中, 还赋予了该框架小样本学习的能 力. 本文以一类典型的量子受限材料——石墨烯纳米带为例, 展示了该 框架的构造过程及其在小样本数据集上训练后实现的优秀的能带结构 预测能力. 我们的机器学习框架提供了一种确定电子能带结构的快速 可靠的方法, 且对图表示在相关领域中的应用具有启发意义.
Source: expertdata · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Machine Learning in Materials Science · Graphene research and applications · Fuel Cells and Related Materials
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (75)
Show source-provided references
1. Carleo G, Cirac I, Cranmer K, et al. Machine learning and the physical sciences. Rev Mod Phys, 2019, 91: 04500210.1103/RevModPhys.91.045002 ↗
2. Dral PO. Quantum chemistry in the age of machine learning. J Phys Chem Lett, 2020, 11: 2336–234710.1021/acs.jpclett.9b03664 ↗
3. Westermayr J, Gastegger M, Schütt KT, et al. Perspective on integrating machine learning into computational chemistry and materials science. J Chem Phys, 2021, 154: 23090310.1063/5.0047760 ↗
4. Ward L, Liu R, Krishna A, et al. Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations. Phys Rev B, 2017, 96: 02410410.1103/PhysRevB.96.024104 ↗
5. Faber FA, Lindmaa A, von Lilienfeld OA, et al. Machine learning energies of 2 million elpasolite (ABC2D6) crystals. Phys Rev Lett, 2016, 117: 13550210.1103/PhysRevLett.117.135502 ↗
6. Ghosh K, Stuke A, Todorović M, et al. Deep learning spectroscopy: Neural networks for molecular excitation spectra. Adv Sci, 2019, 6: 180136710.1002/advs.201801367 ↗
7. Carrete J, Mingo N, Wang S, et al. Nanograined half-Heusler semiconductors as advanced thermoelectrics: An ab initio high-throughput statistical study. Adv Funct Mater, 2014, 24: 7427–743210.1002/adfm.201401201 ↗
8. Ju S, Shiga T, Feng L, et al. Designing nanostructures for phonon transport via Bayesian optimization. Phys Rev X, 2017, 7: 021024
9. Schütt KT, Glawe H, Brockherde F, et al. How to represent crystal structures for machine learning: Towards fast prediction of electronic properties. Phys Rev B, 2014, 89: 20511810.1103/PhysRevB.89.205118 ↗
10. Seko A, Hayashi H, Nakayama K, et al. Representation of compounds for machine-learning prediction of physical properties. Phys Rev B, 2017, 95: 14411010.1103/PhysRevB.95.144110 ↗
11. Xue D, Balachandran PV, Hogden J, et al. Accelerated search for materials with targeted properties by adaptive design. Nat Commun, 2016, 7: 1124110.1038/ncomms11241 ↗
12. Ghiringhelli LM, Vybiral J, Levchenko SV, et al. Big data of materials science: Critical role of the descriptor. Phys Rev Lett, 2015, 114: 10550310.1103/PhysRevLett.114.105503 ↗
13. von Lilienfeld OA, Ramakrishnan R, Rupp M, et al. Fourier series of atomic radial distribution functions: A molecular fingerprint for machine learning models of quantum chemical properties. Int J Quantum Chem, 2015, 115: 1084–109310.1002/qua.24912 ↗
14. Wu Z, Pan S, Chen F, et al. A comprehensive survey on graph neural networks. IEEE Trans Neural Netw Learn Syst, 2021, 32: 4–2410.1109/TNNLS.2020.2978386 ↗
15. Xie T, Grossman JC. Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties. Phys Rev Lett, 2018, 120: 14530110.1103/PhysRevLett.120.145301 ↗
16. Xie T, Grossman JC. Hierarchical visualization of materials space with graph convolutional neural networks. J Chem Phys, 2018, 149: 17411110.1063/1.5047803 ↗
17. Karamad M, Magar R, Shi Y, et al. Orbital graph convolutional neural network for material property prediction. Phys Rev Mater, 2020, 4: 09380110.1103/PhysRevMaterials.4.093801 ↗
18. Korolev V, Mitrofanov A, Korotcov A, et al. Graph convolutional neural networks as “general-purpose” property predictors: The universality and limits of applicability. J Chem Inf Model, 2020, 60: 22–2810.1021/acs.jcim.9b00587 ↗
19. Park CW, Wolverton C. Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery. Phys Rev Mater, 2020, 4: 06380110.1103/PhysRevMaterials.4.063801 ↗
20. Chen C, Zuo Y, Ye W, et al. Learning properties of ordered and disordered materials from multi-fidelity data. Nat Comput Sci, 2021, 1: 46–5310.1038/s43588-020-00002-x ↗
21. Chen C, Ye W, Zuo Y, et al. Graph networks as a universal machine learning framework for molecules and crystals. Chem Mater, 2019, 31: 3564–357210.1021/acs.chemmater.9b01294 ↗
22. Slater JC, Koster GF. Simplified LCAO method for the periodic potential problem. Phys Rev, 1954, 94: 1498–152410.1103/PhysRev.94.1498 ↗
23. Li X, Wang X, Zhang L, et al. Chemically derived, ultrasmooth graphene nanoribbon semiconductors. Science, 2008, 319: 1229–123210.1126/science.1150878 ↗
24. Cai J, Ruffieux P, Jaafar R, et al. Atomically precise bottom-up fabrication of graphene nanoribbons. Nature, 2010, 466: 470–47310.1038/nature09211 ↗
25. Son YW, Cohen ML, Louie SG. Energy gaps in graphene nanoribbons. Phys Rev Lett, 2006, 97: 21680310.1103/PhysRevLett.97.216803 ↗
26. Nakada K, Fujita M, Dresselhaus G, et al. Edge state in graphene ribbons: Nanometer size effect and edge shape dependence. Phys Rev B, 1996, 54: 17954–1796110.1103/PhysRevB.54.17954 ↗
27. Jain A, Ong SP, Hautier G, et al. Commentary: The materials project: A materials genome approach to accelerating materials innovation. APL Mater, 2013, 1: 01100210.1063/1.4812323 ↗
28. Choudhary K, Garrity KF, Reid ACE, et al. The joint automated repository for various integrated simulations (JARVIS) for data-driven materials design. npj Comput Mater, 2020, 6: 17310.1038/s41524-020-00440-1 ↗
29. Lv Y, Ye S, Wang H, et al. Strain engineering of chevron graphene nanoribbons. J Appl Phys, 2018, 125: 08250110.1063/1.5048527 ↗
30. Cai J, Pignedoli CA, Talirz L, et al. Graphene nanoribbon heterojunctions. Nat Nanotech, 2014, 9: 896–90010.1038/nnano.2014.184 ↗
31. Ye S, Lv Y, Tang Z, et al. Wave-function symmetry mechanism of quantum-well states in graphene nanoribbon heterojunctions. Phys Rev Appl, 2019, 12: 04401810.1103/PhysRevApplied.12.044018 ↗
32. Ye S, Wang H, Qiu M, et al. The reconstruction of the symmetry between sublattices: A strategy to improve the transport properties of edge-defective graphene nanoribbon transistors. Phys Chem Chem Phys, 2020, 22: 18265–1827110.1039/D0CP01684E ↗
33. Lv Y, Chang S, Wang H, et al. Energy gap tunable graphene antidot nanoribbon MOSFET: A uniform multiscale analysis from band structure to transport properties. Carbon, 2016, 101: 143–15110.1016/j.carbon.2016.01.091 ↗
34. de Oteyza DG, García-Lekue A, Vilas-Varela M, et al. Substrate-independent growth of atomically precise chiral graphene nanoribbons. ACS Nano, 2016, 10: 9000–900810.1021/acsnano.6b05269 ↗
35. Rizzo DJ, Veber G, Jiang J, et al. Inducing metallicity in graphene nanoribbons via zero-mode superlattices. Science, 2020, 369: 1597–160310.1126/science.aay3588 ↗
36. Lv Y, Huang Q, Chang S, et al. Novel strategy of edge saturation Hamiltonian for graphene nanoribbon devices. IEEE Trans Electron Devices, 2016, 63: 4514–452010.1109/TED.2016.2603156 ↗
37. Ye S, Wang Z, Wang H, et al. Steep-slope transistors based on chiral graphene nanoribbons with intrinsic cold source. IEEE Trans Electron Devices, 2021, 68: 4123–412810.1109/TED.2021.3087459 ↗
38. Giannozzi P, Baroni S, Bonini N, et al. Quantum espresso: A modular and open-source software project for quantum simulations of materials. J Phys-Condens Matter, 2009, 21: 39550210.1088/0953-8984/21/39/395502 ↗
39. Monkhorst HJ, Pack JD. Special points for Brillouin-zone integrations. Phys Rev B, 1976, 13: 5188–519210.1103/PhysRevB.13.5188 ↗
40. Perdew JP, Burke K, Ernzerhof M. Generalized gradient approximation made simple. Phys Rev Lett, 1996, 77: 386510.1103/PhysRevLett.77.3865 ↗
41. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature, 2015, 521: 436–44410.1038/nature14539 ↗
42. Unke OT, Meuwly M. Physnet: A neural network for predicting energies, forces, dipole moments, and partial charges. J Chem Theor Comput, 2019, 15: 3678–369310.1021/acs.jctc.9b00181 ↗
43. Gong S, Xie T, Zhu T, et al. Predicting charge density distribution of materials using a local-environment-based graph convolutional network. Phys Rev B, 2019, 100: 18410310.1103/PhysRevB.100.184103 ↗
44. Gasteiger J, Groß J, Günnemann S. Directional message passing for molecular graphs. In: Proceedings of the 8th International Conference on Learning Representations (ICLR). Addis Ababa, 2020
45. Zhou J, Cui G, Hu S, et al. Graph neural networks: A review of methods and applications. AI Open, 2020, 1: 57–8110.1016/j.aiopen.2021.01.001 ↗
46. Gilmer J, Schoenholz SS, Riley PF, et al. Neural message passing for quantum chemistry. In: Proceedings of the 34th International Conference on Machine Learning (ICML). Sydney, 2017. 1263–1272
47. Veličković P, Cucurull G, Casanova A, et al. Graph attention networks. In: Proceedings of the 6th International Conference on Learning Representations (ICLR). Vancouver, 2018
48. Nair V, Hinton G. Rectified linear units improve restricted Boltzmann machines. In: Proceedings of the 27th International Conference on Machine Learning (ICML). Haifa, 2010
49. He K, Zhang X, Ren S, et al. Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: Proceedings of the 15th International Conference on Computer Vision (ICCV). Santiago, 2015. 1026103410.1109/ICCV.2015.123 ↗
50. Goringe CM, Bowler DR, Hernández E. Tight-binding modelling of materials. Rep Prog Phys, 1997, 60: 1447–151210.1088/0034-4885/60/12/001 ↗
51. Kohn W, Sham LJ. Self-consistent equations including exchange and correlation effects. Phys Rev, 1965, 140: A1133–A113810.1103/PhysRev.140.A1133 ↗
52. Andrade X, Aspuru-Guzik A. Real-space density functional theory on graphical processing units: Computational approach and comparison to Gaussian basis set methods. J Chem Theor Comput, 2013, 9: 4360–437310.1021/ct400520e ↗
53. Kohn W. Density functional and density matrix method scaling linearly with the number of atoms. Phys Rev Lett, 1996, 76: 3168–317110.1103/PhysRevLett.76.3168 ↗
54. Harrison WA. Electronic structure and the properties of interfaces. Ultramicroscopy, 1984, 14: 85–8710.1016/0304-3991(84)90111-6 ↗
55. Girshick R. Fast R-CNN. In: Proceedings of the 15th International Conference on Computer Vision (ICCV). Santiago, 2015. 1440–144810.1109/ICCV.2015.169 ↗
56. Kingma DP, Ba J. Adam: A method for stochastic optimization. In: Proceedings of the 3rd International Conference on Learning Representations (ICLR). San Diego, 2015
57. Gasteiger J, Giri S, Margraf JT, et al. Fast and uncertainty-aware directional message passing for non-equilibrium molecules. In: Proceedings of the 34th Conference and Workshop on Neural Information Processing Systems (NeurIPS). Vancouver, 2020
58. Choudhary K, DeCost B. Atomistic line graph neural network for improved materials property predictions. npj Comput Mater, 2021, 7: 18510.1038/s41524-021-00650-1 ↗
59. Soler JM, Artacho E, Gale JD, et al. The SIESTA method for ab initio order-N materials simulation. J Phys-Condens Matter, 2002, 14: 2745–277910.1088/0953-8984/14/11/302 ↗
60. Köhler C, Frauenheim T. Molecular dynamics simulations of CFx (x = 2, 3) molecules at Si3N4 and SiO2 surfaces. Surf Sci, 2006, 600: 453–46010.1016/j.susc.2005.10.044 ↗
61. Smidstrup S, Markussen T, Vancraeyveld P, et al. Quantumatk: An integrated platform of electronic and atomic-scale modelling tools. J Phys-Condens Matter, 2019, 32: 01590110.1088/1361-648X/ab4007 ↗
62. Wang Y, Yao Q, Kwok JT, et al. Generalizing from a few examples. ACM Comput Surv, 2021, 53: 1–3410.1145/3386252 ↗
63. Rudenko AN, Yuan S, Katsnelson MI. Toward a realistic description of multilayer black phosphorus: From GW approximation to large-scale tight-binding simulations. Phys Rev B, 2015, 92: 08541910.1103/PhysRevB.92.085419 ↗
64. Cappelluti E, Roldán R, Silva-Guillén JA, et al. Tight-binding model and direct-gap/indirect-gap transition in single-layer and multilayer MoS2. Phys Rev B, 2013, 88: 07540910.1103/PhysRevB.88.075409 ↗
65. Fiori G, Iannaccone G. Multiscale modeling for graphene-based nanoscale transistors. Proc IEEE, 2013, 101: 1653–166910.1109/JPROC.2013.2259451 ↗
66. Wang Z, Ye S, Wang H, et al. Machine learning method for tight-binding Hamiltonian parameterization from ab-initio band structure. npj Comput Mater, 2021, 7: 1110.1038/s41524-020-00490-5 ↗
67. Datta S. Nanoscale device modeling: The green’s function method. Superlat Microstruct, 2000, 28: 253–27810.1006/spmi.2000.0920 ↗
68. Lee DH, Joannopoulos JD. Simple scheme for surface-band calculations. I. Phys Rev B, 1981, 23: 4988–499610.1103/PhysRevB.23.4988 ↗
69. Marzari N, Vanderbilt D. Maximally localized generalized Wannier functions for composite energy bands. Phys Rev B, 1997, 56: 12847–1286510.1103/PhysRevB.56.12847 ↗
70. Souza I, Marzari N, Vanderbilt D. Maximally localized Wannier functions for entangled energy bands. Phys Rev B, 2001, 65: 03510910.1103/PhysRevB.65.035109 ↗
71. Marzari N, Mostofi AA, Yates JR, et al. Maximally localized Wannier functions: Theory and applications. Rev Mod Phys, 2012, 84: 1419–147510.1103/RevModPhys.84.1419 ↗
72. Stieger C, Szabo A, Bunjaku T, et al. Ab-initio quantum transport simulation of self-heating in single-layer 2-D materials. J Appl Phys, 2017, 122: 04570810.1063/1.4990384 ↗
73. Marin EG, Marian D, Iannaccone G, et al. First-principles simulations of FETs based on two-dimensional InSe. IEEE Electron Device Lett, 2018, 39: 626–62910.1109/LED.2018.2804388 ↗
74. Marin EG, Marian D, Iannaccone G, et al. First principles investigation of tunnel FETs based on nanoribbons from topological two-dimensional materials. Nanoscale, 2017, 9: 19390–1939710.1039/C7NR06015G ↗
75. Fiori G, Iannaccone G. Simulation of graphene nanoribbon field-effect transistors. IEEE Electron Device Lett, 2007, 28: 760–76210.1109/LED.2007.901680 ↗