Chinese Journal of Catalysis ›› 2022, Vol. 43 ›› Issue (1): 11-32.DOI: 10.1016/S1872-2067(21)63852-4
• Reviews • Previous Articles Next Articles
Letian Chena, Xu Zhanga,b,#(), An Chena, Sai Yaoa, Xu Hua, Zhen Zhoua,b,*()
Received:
2021-05-11
Accepted:
2021-05-28
Online:
2022-01-18
Published:
2021-07-07
Contact:
Xu Zhang,Zhen Zhou
About author:
# E-mail: zhangxu@nankai.edu.cnSupported by:
Letian Chen, Xu Zhang, An Chen, Sai Yao, Xu Hu, Zhen Zhou. Targeted design of advanced electrocatalysts by machine learning[J]. Chinese Journal of Catalysis, 2022, 43(1): 11-32.
Add to citation manager EndNote|Ris|BibTeX
URL: https://www.cjcatal.com/EN/10.1016/S1872-2067(21)63852-4
Fig. 2. (a) Schematic illustration of the procedure for prediction of new compounds by using DMSP. Adapted with permission from Ref. [17]. Copyright 2011, American Chemical Society. (b) Discovery histogram for new ternary nitrides based on entries catalogued in ICSD. Adapted with permission from Ref. [51]. Copyright 2019, Nature.
Fig. 3. (a) Breeding processes in GAs for two example parents Ni6@Pt32 and Co6@Pt32. Adapted with permission from Ref. [19]. Copyright 2009, AIP publishers. (b) A schematic plot of generative machine learning model based on a generative adversarial network for inorganic materials. Adapted with permission from Ref. [22]. Copyright 2020, the Authors.
Fig. 4. (a) The confusion matrix with the F-Measure values of classification models. Adapted with permission from Ref. [49]. Copyright 2020, the Authors. (b) The convex hull located by ML-based GA. Adapted with permission from Ref. [23]. Copyright 2019, the Authors.
Fig. 5. (a) Relationship between the descriptor found by LASSO and overpotential of OER. Adapted with permission from Ref. [72]. Copyright 2020, American Chemical Society. (b) Relationship between the descriptor found by LASSO and Gibbs free energy of first hydrogenation for NRR. Inset: the optimized structures for N2 on 2D Ti@VB2. Adapted with permission from Ref. [73]. Copyright 2020, American Chemical Society. CO adsorption energies calculated by DFT vs. the one predicted by (c) the two-level interaction model or (d) ML model. Insets: sketches of (c) the two-level interaction model and (d) the artificial neural network models. (c,d) Adapted with permission from Ref. [74]. Copyright 2015, American Chemical Society. Comparison of the performance of ML for predicting adsorption energies trained with (e) the geometry based and (f) electronic structured based primary features. (e,f) Adapted with permission from Ref. [75]. Copyright 2017, Elsevier.
Fig. 6. (a) The limiting potential steps for MOR plotted with CO and OH adsorption free energies relative to Pt(111). Adapted with permission from Ref. [81]. Copyright 2017, Royal Society of Chemistry. (b) Performance of the GBR model for predicting adsorption energies of different adsorbates: CO (left), CHx (middle), and CO/CHO/COH (right). Insets: the feature importance scores of different descriptors (upper left corner) and the prediction error distribution (lower right corner). (c) Performance of the GBR model for predicting adsorption energies of adsorbates (11 species). Insets: the feature importance scores of different descriptors and the prediction error distribution. (b,c) Adapted with permission from Ref. [83]. Copyright 2017, Royal Society of Chemistry.
Fig. 7. (a) Heat map of the prediction values for CO adsorption energies. Adapted with permission from Ref. [85]. Copyright 2020, American Chemical Society. (b) t-SNE visualization of approximately 4000 adsorption sites simulated with DFT. (c) The sites for each cluster labelled in (b). (d) 2D selectivity volcano plot for CO2RR [91]. Copyright 2020, Nature. (e) The adsorption Gibbs free energies of N2 versus the adsorption Gibbs free energies of H for 23 transition metals. (b-e) Adapted with permission from Ref. [26]. Copyright 2020, American Chemical Society. (f) Visualization of catalytic sites on dealloyed gold surface. Adapted with permission from Ref. [92]. Copyright 2019, American Chemical Society.
Fig. 8. (a) Distribution of OH adsorption energies for 871 cells. Each individual on-top binding site is represented by a color. Adapted with permission from Ref. [25]. Copyright 2019, Elsevier. (b) iR-corrected LSV curve of Al0.5Mn2.5O4. Inset: the volcano plot where the location of Al0.5Mn2.5O4 is. The volcano plot represents the relationship between the experimentally observed reaction activity and the calculated Max(DT, DO), where DT is the distance between O p-band center and tetrahedral metal d-band center, and DO represents the distance from O p-band center to octahedral metal d-band center. Adapted with permission from Ref. [104]. Copyright 2019, Nature.
Fig. 9. Schematic diagram of (a) the FFNN for fitting PES and (b) the HDNN framework for a ternary system with A, B and C elements. (a,b) Adapted with permission from Ref. [38]. Copyright 2017, Wiley-VCH. (c) The process for the fitting of the PES by HDNN. Adapted with permission from Ref. [121]. Copyright 2015, American Chemical Society. (d) H-coupling reaction path of H2 production on amorphous surface of TiO2-0.69H. Adapted with permission from Ref. [39]. Copyright 2018, American Chemical Society. (e) Phase diagram of ternary Zn-Cr-O. Adapted with permission from Ref. [122]. Copyright 2019, Nature. (f) Reaction path of the acetylene hydrogenation on Pd(111) and Pd4H3(111). Adapted with permission from Ref. [123]. Copyright 2020, American Chemical Society.
Fig. 10. (a) Reaction network for the reaction of syngas to CO2, H2O, CH3OH, CH3CHO, CH4, CH3CH2OH and other intermediates. (b) The structure of Rh(111) and the reduced reaction network for syngas reaction on Rh(111) surface. The cyan, red, grey and white spheres represent Rh, O, C, and H atoms, respectively. Adapted with permission from Ref. [36]. Copyright 2017, the Authors.
Fig. 11. (a) Schematic of the BR used to predict energetics of reaction intermediates on nanoparticles. The training data of adsorbate binding energies on single-crystal surfaces and small clusters were calculated by DFT. Ek, Kkj, and wj represent the energy of kth reaction intermediate on nanoparticles, the SOAP kernel, and the regression coefficients, respectively [37]. Copyright 2017, American Chemical Society. (b) Predicted TOFs of each surface site at 500 K. Inset: a structure of the active site on Rh(1-x)Aux. Adapted with permission from Ref. [130]. Copyright 2017, American Chemical Society.
Fig. 12. (a) Schematic of ReQM. (b) Distribution of the formation energies for HOCO (left) and CO (right). Adapted with permission from Ref. [140]. Copyright 2021, Elsevier.
Fig. 13. (a) The electronic energy E and corresponding work function Φ as a function of reaction coordinates. (b) Reaction and activation energies calculated at different cell sizes ai vs. the change in potential and extrapolation of reaction and activation energies from unit cell size a0 to infinite cell size a∞. (c) Parity plot of reaction energies and barriers obtained by cell-extrapolation and the charge-extrapolation. Adapted with permission from Ref [158]. Copyright 2015, American Chemical Society.
Fig. 14. The prospect of machine learning in the explicit solvent model under constant potentials. The right part is the atomic structure of an electrocatalyst in an aqueous solution, including a porous carbon electrode, electrified interface and electrolyte solution. Adapted with permission from Ref [162]. Copyright 2021, Nature.
|
[1] | Bo Zhou, Jianqiao Shi, Yimin Jiang, Lei Xiao, Yuxuan Lu, Fan Dong, Chen Chen, Tehua Wang, Shuangyin Wang, Yuqin Zou. Enhanced dehydrogenation kinetics for ascorbic acid electrooxidation with ultra-low cell voltage and large current density [J]. Chinese Journal of Catalysis, 2023, 50(7): 372-380. |
[2] | Ling Ouyang, Jie Liang, Yongsong Luo, Dongdong Zheng, Shengjun Sun, Qian Liu, Mohamed S. Hamdy, Xuping Sun, Binwu Ying. Recent advances in electrocatalytic ammonia synthesis [J]. Chinese Journal of Catalysis, 2023, 50(7): 6-44. |
[3] | Yuzhuo Chen, Hao Wang, Bing Lu, Ni Yi, Liang Cao, Yong Wang, Shanjun Mao. Fine-structure sensitive deep learning framework for predicting catalytic properties with high precision [J]. Chinese Journal of Catalysis, 2023, 50(7): 284-296. |
[4] | Shuaiqi Meng, Zhongyu Li, Peng Zhang, Francisca Contreras, Yu Ji, Ulrich Schwaneberg. Deep learning guided enzyme engineering of Thermobifida fusca cutinase for increased PET depolymerization [J]. Chinese Journal of Catalysis, 2023, 50(7): 229-238. |
[5] | Jingjing Li, Fengwei Zhang, Xinyu Zhan, Hefang Guo, Han Zhang, Wen-Yan Zan, Zhenyu Sun, Xian-Ming Zhang. Precise design of nickel phthalocyanine molecular structure: Optimizing electronic and spatial effects for remarkable electrocatalytic CO2 reduction [J]. Chinese Journal of Catalysis, 2023, 48(5): 117-126. |
[6] | Wenjing Zhang, Jing Li, Zidong Wei. Carbon-based catalysts of the oxygen reduction reaction: Mechanistic understanding and porous structures [J]. Chinese Journal of Catalysis, 2023, 48(5): 15-31. |
[7] | Zexing Wu, Yuxiao Gao, Zixuan Wang, Weiping Xiao, Xinping Wang, Bin Li, Zhenjiang Li, Xiaobin Liu, Tianyi Ma, Lei Wang. Surface-enriched ultrafine Pt nanoparticles coupled with defective CoP as efficient trifunctional electrocatalyst for overall water splitting and flexible Zn-air battery [J]. Chinese Journal of Catalysis, 2023, 46(3): 36-47. |
[8] | Xiaoni Liu, Xiaobin Liu, Caixia Li, Bo Yang, Lei Wang. Defect engineering of electrocatalysts for metal-based battery [J]. Chinese Journal of Catalysis, 2023, 45(2): 27-87. |
[9] | Lijing Wang, Tianyi Yang, Bo Feng, Xiangyu Xu, Yuying Shen, Zihan Li, Arramel , Jizhou Jiang. Constructing dual electron transfer channels to accelerate CO2 photoreduction guided by machine learning and first-principles calculation [J]. Chinese Journal of Catalysis, 2023, 54(11): 265-277. |
[10] | Yuxuan Lu, Liu Yang, Yimin Jiang, Zhenran Yuan, Shuangyin Wang, Yuqin Zou. Engineering a localized electrostatic environment to enhance hydroxyl activating for electrocatalytic biomass conversion [J]. Chinese Journal of Catalysis, 2023, 53(10): 153-160. |
[11] | Yuyan Qiao, Yanqiu Pan, Jiangwei Zhang, Bin Wang, Tingting Wu, Wenjun Fan, Yucheng Cao, Rashid Mehmood, Fei Zhang, Fuxiang Zhang. Multiple carbon interface engineering to boost oxygen evolution of NiFe nanocomposite electrocatalyst [J]. Chinese Journal of Catalysis, 2022, 43(9): 2354-2362. |
[12] | Xiao-Tian Li, Lin Chen, Cheng Shang, Zhi-Pan Liu. Selectivity control in alkyne semihydrogenation: Recent experimental and theoretical progress [J]. Chinese Journal of Catalysis, 2022, 43(8): 1991-2000. |
[13] | Tong Cui, Xuejun Zhai, Lili Guo, Jing-Qi Chi, Yu Zhang, Jiawei Zhu, Xuemei Sun, Lei Wang. Controllable synthesis of a self-assembled ultralow Ru, Ni-doped Fe2O3 lily as a bifunctional electrocatalyst for large-current-density alkaline seawater electrolysis [J]. Chinese Journal of Catalysis, 2022, 43(8): 2202-2211. |
[14] | Xiu Qian, Yanjiao Wei, Mengjie Sun, Ye Han, Xiaoli Zhang, Jian Tian, Minhua Shao. Heterostructuring 2D TiO2 nanosheets in situ grown on Ti3C2Tx MXene to improve the electrocatalytic nitrogen reduction [J]. Chinese Journal of Catalysis, 2022, 43(7): 1937-1944. |
[15] | Lili Zhang, Suyu Jiang, Wei Ma, Zhen Zhou. Oxygen reduction reaction on Pt-based electrocatalysts: Four-electron vs. two-electron pathway [J]. Chinese Journal of Catalysis, 2022, 43(6): 1433-1443. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||