Citation: Chen Chen, Jinzhou Zheng, Chaoqin Chu, Qinkun Xiao, Chaozheng He, Xi Fu. An effective method for generating crystal structures based on the variational autoencoder and the diffusion model[J]. Chinese Chemical Letters, ;2025, 36(4): 109739. doi: 10.1016/j.cclet.2024.109739 shu

An effective method for generating crystal structures based on the variational autoencoder and the diffusion model

Chen Chen ^{a, 1} ,
Jinzhou Zheng ^{b, 1} ,
Chaoqin Chu ^{c, 1} ,
Qinkun Xiao ^{a, b, *,} ,
Chaozheng He ^{b, *,} ,
Xi Fu ^d

a.
School of Electrical and Information Engineering, Xi’an Technological University, Xi’an 710021, China
b.
School of Materials Science and Chemical Engineering, Xi’an Technological University, Xi’an 710021, China
c.
School of Mechatronic Engineering, Xi’an Technological University, Xi’an 710021, China
d.
College of Science, Hunan Universtiy of Science and Engineering, Yongzhou 425199, China

xiaoqinkun10000@163.com

hecz2019@xatu.edu.cn

Received Date: 23 February 2024
Revised Date: 6 March 2024
Accepted Date: 6 March 2024
Available Online: 8 March 2024

Figures(6)

Two dimensional (2D) materials based on boron and carbon have attracted wide attention due to their unique properties. BC compounds have rich active sites and diverse chemical coordination, showing great potential in optoelectronic applications. However, due to the limitation of calculation and experimental conditions, it is still a challenging task to predict new 2D BC monolayer materials. Specifically, we utilized Crystal Diffusion Variational Autoencoder (CDVAE) and pre-trained Materials Graph Neural Network with 3-Body Interactions (M3GNet) model to generate novel and stable BCP materials. Each crystal structure was treated as a high-dimensional vector, where the encoder extracted lattice information and element coordinates, mapping the high-dimensional data into a low-dimensional latent space. The decoder then reconstructed the latent representation back into the original data space. Additionally, our designed attribute predictor network combined the advantages of dilated convolutions and residual connections, effectively increasing the model's receptive field and learning capacity while maintaining relatively low parameter count and computational complexity. By progressively increasing the dilation rate, the model can capture features at different scales. We used the DFT data set of about 1600 BCP monolayer materials to train the diffusion model, and combined with the pre-trained M3GNet model to screen the best candidate structure. Finally, we used DFT calculations to confirm the stability of the candidate structure. The results show that the combination of generative deep learning model and attribute prediction model can help accelerate the discovery and research of new 2D materials, and provide effective methods for exploring the inverse design of new two-dimensional materials.
- Deep generative model,
- BCP monolayer,
- Inverse design,
- CDVAE,
- DFT
1. [1]
  X. Fu, X. Cheng, C. He, et al., Phys. Chem. Chem. Phys. 25 (3) (2023) 2430–2438. doi: 10.1039/d2cp04941d
2. [2]
  M. Turunen, M. Brotons-Gisbert, Y. Dai, et al., Nat. Rev. Phys. 4 (4) (2022) 219–236. doi: 10.1038/s42254-021-00408-0
3. [3]
  A.R. Khan, L. Zhang, K. Ishfaq, et al., Adv. Funct. Mater. 32 (2022) 2105259. doi: 10.1002/adfm.202105259
4. [4]
  Y. Mo, L. Liao, D. Li, et al., Chin. Chem. Lett. 34 (2023) 107130. doi: 10.1016/j.cclet.2022.01.023
5. [5]
  B. Ryu, L. Wang, H. Pu, et al., Chem. Soc. Rev. 51 (2022) 1899–1925. doi: 10.1039/d1cs00503k
6. [6]
  Y. Song, E.M.D. Siriwardane, Y. Zhao, et al., ACS Appl. Mater. Interfaces 13 (2021) 53303–53313. doi: 10.1021/acsami.1c01044
7. [7]
  L. Shen, J. Zhou, T. Yang, et al., Acc. Mater. Res. 3 (2022) 572–583. doi: 10.1021/accountsmr.1c00246
8. [8]
  P. Lyngby, K.S. Thygesen, Npj Comput. Mater. 8 (2022) 232. doi: 10.1038/s41524-022-00923-3
9. [9]
  K.M. Wyss, D.X. Luong, J.M. Tour, Adv. Mater. 34 (2022) 2106970. doi: 10.1002/adma.202106970
10. [10]
  P. Ares, K.S. Novoselov, Nano Mater. Sci. 4 (2022) 3–9.
11. [11]
  Y. Zhao, M. Al-Fahdi, M. Hu, et al., Adv. Sci. 8 (2021) 2100566. doi: 10.1002/advs.202100566
12. [12]
  C. Chu, Q. Xiao, C. He, et al., Chin. Chem. Lett. 35 (2024) 109186. doi: 10.1016/j.cclet.2023.109186
13. [13]
  Z. Ren, S.I.P. Tian, J. Noh, et al., Matter 5 (2022) 314–335. doi: 10.1016/j.matt.2021.11.032
14. [14]
  T. Yang, J. Zhou, T.T. Song, et al., ACS Energy Lett. 5 (2020) 2313–2321. doi: 10.1021/acsenergylett.0c00957
15. [15]
  S. Lu, Q. Zhou, Y. Guo, et al., Adv. Mater. 32 (2020) 2002658. doi: 10.1002/adma.202002658
16. [16]
  Y. Zuo, M. Qin, C. Chen, et al., Mater. Today 51 (2021) 126–135. doi: 10.1016/j.mattod.2021.08.012
17. [17]
  Y. Chen, J. Zhu, Y. Xie, et al., Nanoscale 11 (2019) 9749–9755. doi: 10.1039/c9nr01315f
18. [18]
  M.C. Sorkun, S. Astruc, J.V.A. Koelman, et al., Npj Comput. Mater. 6 (2020) 106. doi: 10.1038/s41524-020-00375-7
19. [19]
  S. Manti, M.K. Svendsen, N.R. Knøsgaard, et al., Npj Comput. Mater. 9 (2023) 33. doi: 10.1038/s41524-023-00977-x
20. [20]
  A. Karthikeyan, U.D. Priyakumar, J. Chem. Sci. 134 (2022) 1–20. doi: 10.1007/s12039-021-01987-2
21. [21]
  T. Xie, X. Fu, O.E. Ganea, et al., arXiv (2021) arXiv:2110.06197.
22. [22]
  C. Chen, S.P. Ong, Nat. Comput. Sci. 2 (2022) 718–728. doi: 10.1038/s43588-022-00349-3
23. [23]
  D.P. Kingma, M. Welling, arXiv (2013) arXiv:1312.6114.
24. [24]
  J. Ho, A. Jain, P. Abbeel, Adv. Neural. Inf. Process Syst. 33 (2020) 6840–6851.
25. [25]
  J. Han, Y. Rong, T. Xu, et al., arXiv (2022) arXiv:2202.07230.
26. [26]
  Y. Song, S. Ermon, Adv. Neural. Inf. Process. Syst. 32 (2019) 11918–11930.
27. [27]
  D.P. Kingma, J. Ba, arXiv (2014) arXiv:1412.6980.
28. [28]
  C.J. Court, B. Yildirim, A. Jain, et al., J. Chem. Inf. Model. 60 (2020) 4518–4535. doi: 10.1021/acs.jcim.0c00464
29. [29]
  H. Moustafa, P.M. Lyngby, J.J. Mortensen, et al., Phys. Rev. Mater. 7 (2023) 014007. doi: 10.1103/PhysRevMaterials.7.014007
30. [30]
  M.N. Gjerding, A. Taghizadeh, A. Rasmussen, et al., 2D Mater. 8 (2021) 044002.
31. [31]
  T. Xie, J.C. Grossman, Phys. Rev. Lett. 120 (2018) 145301.
32. [32]
  X. Fu, J. Guo, L. Li, et al., Chem. Phys. Lett. 726 (2019) 69–76.
33. [33]
  A. Togo, F. Oba, I. Tanaka, Phys. Rev. B 78 (2008) 134106. doi: 10.1103/PhysRevB.78.134106
34. [34]
  A. Bafekry, M. Naseri, M. Faraji, et al., Sci. Rep. 12 (2022) 22269.
35. [35]
  M. Tang, C. Wang, U. Schwingenschlögl, et al., Chem. Mater. 33 (2021) 9262–9269. doi: 10.1021/acs.chemmater.1c02957
36. [36]
  R. Muthaiah, F. Tarannum, R.S. Annam, et al., RSC Adv. 10 (2020) 42628–42632. doi: 10.1039/d0ra08444a
1. [1]
  Zhengyu Zhou , Huiqin Yao , Youlin Wu , Teng Li , Noritatsu Tsubaki , Zhiliang Jin . Synergistic Effect of Cu-Graphdiyne/Transition Bimetallic Tungstate Formed S-Scheme Heterojunction for Enhanced Photocatalytic Hydrogen Evolution. Acta Physico-Chimica Sinica, 2024, 40(10): 2312010-. doi: 10.3866/PKU.WHXB202312010
2. [2]
  Tsegaye Tadesse Tsega , Jiantao Zai , Chin Wei Lai , Xin-Hao Li , Xuefeng Qian . Earth-abundant CuFeS2 nanocrystals@graphite felt electrode for high performance aqueous polysulfide/iodide redox flow batteries. Chinese Journal of Structural Chemistry, 2024, 43(1): 100192-100192. doi: 10.1016/j.cjsc.2023.100192
3. [3]
  Chunyan Yang , Qiuyu Rong , Fengyin Shi , Menghan Cao , Guie Li , Yanjun Xin , Wen Zhang , Guangshan Zhang . Rationally designed S-scheme heterojunction of BiOCl/g-C₃N₄ for photodegradation of sulfamerazine: Mechanism insights, degradation pathways and DFT calculation. Chinese Chemical Letters, 2024, 35(12): 109767-. doi: 10.1016/j.cclet.2024.109767
4. [4]
  Run-Han Li , Tian-Yi Dang , Wei Guan , Jiang Liu , Ya-Qian Lan , Zhong-Min Su . Evolution exploration and structure prediction of Keggin-type group IVB metal-oxo clusters. Chinese Chemical Letters, 2024, 35(5): 108805-. doi: 10.1016/j.cclet.2023.108805
5. [5]
  Chaozheng He , Jia Wang , Ling Fu , Wei Wei . Nitric oxide assists nitrogen reduction reaction on 2D MBene: A theoretical study. Chinese Chemical Letters, 2024, 35(5): 109037-. doi: 10.1016/j.cclet.2023.109037
6. [6]
  Ting-Ting Huang , Jin-Fa Chen , Juan Liu , Tai-Bao Wei , Hong Yao , Bingbing Shi , Qi Lin . A novel fused bi-macrocyclic host for sensitive detection of Cr₂O₇²⁻ based on enrichment effect. Chinese Chemical Letters, 2024, 35(7): 109281-. doi: 10.1016/j.cclet.2023.109281
7. [7]
  Sanmei Wang , Yong Zhou , Hengxin Fang , Chunyang Nie , Chang Q Sun , Biao Wang . Constant-potential simulation of electrocatalytic N₂ reduction over atomic metal-N-graphene catalysts. Chinese Chemical Letters, 2025, 36(3): 110476-. doi: 10.1016/j.cclet.2024.110476
8. [8]
  Sanmei Wang , Dengxin Yan , Wenhua Zhang , Liangbing Wang . Graphene-supported isolated platinum atoms and platinum dimers for CO₂ hydrogenation: Catalytic activity and selectivity variations. Chinese Chemical Letters, 2025, 36(4): 110611-. doi: 10.1016/j.cclet.2024.110611
9. [9]
  Xin-Tong Zhao , Jin-Zhi Guo , Wen-Liang Li , Jing-Ping Zhang , Xing-Long Wu . Two-dimensional conjugated coordination polymer monolayer as anode material for lithium-ion batteries: A DFT study. Chinese Chemical Letters, 2024, 35(6): 108715-. doi: 10.1016/j.cclet.2023.108715
10. [10]
  Dan-Ying Xing , Xiao-Dan Zhao , Chuan-Shu He , Bo Lai . Kinetic study and DFT calculation on the tetracycline abatement by peracetic acid. Chinese Chemical Letters, 2024, 35(9): 109436-. doi: 10.1016/j.cclet.2023.109436
11. [11]
  Tao Ban , Xi-Yang Yu , Hai-Kuo Tian , Zheng-Qing Huang , Chun-Ran Chang . One-step conversion of methane and formaldehyde to ethanol over SA-FLP dual-active-site catalysts: A DFT study. Chinese Chemical Letters, 2024, 35(4): 108549-. doi: 10.1016/j.cclet.2023.108549
12. [12]
  Yiwen Xu , Chaozheng He , Chenxu Zhao , Ling Fu . Single-atom Ti doping on S-vacancy two-dimensional CrS₂ as a catalyst for ammonia synthesis: A DFT study. Chinese Chemical Letters, 2025, 36(4): 109797-. doi: 10.1016/j.cclet.2024.109797
13. [13]
  Rui Liu , Jinbo Pang , Weijia Zhou . Monolayer water shepherding supertight MXene/graphene composite films. Chinese Journal of Structural Chemistry, 2024, 43(10): 100329-100329. doi: 10.1016/j.cjsc.2024.100329
14. [14]
  Na Wang , Wang Luo , Huaiyi Shen , Huakai Li , Zejiang Xu , Zhiyuan Yue , Chao Shi , Hengyun Ye , Leping Miao . Crystal engineering regulation achieving inverse temperature symmetry breaking ferroelasticity in a cationic displacement type hybrid perovskite system. Chinese Chemical Letters, 2024, 35(5): 108696-. doi: 10.1016/j.cclet.2023.108696
15. [15]
  Lang Gao , Cen Zhou , Rui Wang , Feng Lan , Bohang An , Xiaozhou Huang , Xiao Zhang . Unveiling inverse vulcanized polymers as metal-free, visible-light-driven photocatalysts for cross-coupling reactions. Chinese Chemical Letters, 2024, 35(4): 108832-. doi: 10.1016/j.cclet.2023.108832
16. [16]
  Yujuan Zhao , Zaiwang Zhao . Monolayer mesoporous nanosheets with surface asymmetry via a dual-emulsion-directed monomicelle assembly. Chinese Journal of Structural Chemistry, 2024, 43(2): 100238-100238. doi: 10.1016/j.cjsc.2024.100238
17. [17]
  Jiayi Lu , Yizhang Li , Hao Jiang , Zhiwen Zhu , Fengru Zheng , Qiang Sun . Preparing sub-monolayer metals with continuous coverage spread for high-throughput growth of metal-organic frameworks. Chinese Chemical Letters, 2025, 36(3): 110394-. doi: 10.1016/j.cclet.2024.110394
18. [18]
  Xiaoli Deng , Xiangchao Lu , Yang Cao , Qianjin Chen . Electrochemical imaging uncovers the heterogeneity of HER activity by sulfur vacancies in molybdenum disulfide monolayer. Chinese Chemical Letters, 2025, 36(3): 110379-. doi: 10.1016/j.cclet.2024.110379
19. [19]
  Yongheng Ren , Yang Chen , Hongwei Chen , Lu Zhang , Jiangfeng Yang , Qi Shi , Lin-Bing Sun , Jinping Li , Libo Li . Electrostatically driven kinetic Inverse CO2/C2H2 separation in LTA-type zeolites. Chinese Journal of Structural Chemistry, 2024, 43(10): 100394-100394. doi: 10.1016/j.cjsc.2024.100394
20. [20]
  Bingwei Wang , Yihong Ding , Xiao Tian . Benchmarking model chemistry composite calculations for vertical ionization potential of molecular systems. Chinese Chemical Letters, 2025, 36(2): 109721-. doi: 10.1016/j.cclet.2024.109721

Get Citation

PDF

XML

Metrics

PDF Downloads(1)
Abstract views(63)
HTML views(3)

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142