化工学报 ›› 2021, Vol. 72 ›› Issue (12): 6262-6273.doi: 10.11949/0438-1157.20211258

• 催化、动力学与反应器 • 上一篇    下一篇

基于机器学习的二氧化碳电化学还原制备甲酸盐研究

刘文萱(),张嘉毅,陆奇,张皓晨()   

  1. 清华大学化学工程系,化学工程联合国家重点实验室,北京 100084
  • 收稿日期:2021-08-31 修回日期:2021-10-28 出版日期:2021-12-05 发布日期:2021-12-22
  • 通讯作者: 张皓晨 E-mail:liuwx21@mails.tsinghua.edu.cn;zhc17@mails.tsinghua.edu.cn
  • 作者简介:刘文萱(1999—),女,博士研究生,liuwx21@mails.tsinghua.edu.cn
  • 基金资助:
    国家重点研发计划项目(2017YFB0702800)

Investigation of electroreduction of carbon dioxide into formate based on machine learning

Wenxuan LIU(),Jiayi ZHANG,Qi LU,Haochen ZHANG()   

  1. Department of Chemical Engineering, State Key Laboratory of Chemical Engineering, Tsinghua University, Beijing 100084, China
  • Received:2021-08-31 Revised:2021-10-28 Published:2021-12-05 Online:2021-12-22
  • Contact: Haochen ZHANG E-mail:liuwx21@mails.tsinghua.edu.cn;zhc17@mails.tsinghua.edu.cn

摘要:

系统研究了不同双金属中心催化剂催化二氧化碳电化学还原制备甲酸盐。借助机器学习,确定了反应中心金属原子序数、电负性和电离能等特征对双金属中心催化剂表面二氧化碳还原具有主要的影响。基于这些特征,通过高通量机器学习快速预测了105种双金属中心催化剂二氧化碳电还原制甲酸盐及其主要竞争反应的Gibbs自由能变,筛选出29种双金属中心催化剂更倾向于二氧化碳还原得到甲酸盐,是潜在的转化二氧化碳为甲酸盐的高性能催化材料。运用类似的方法预测了105种双金属中心催化剂表面二氧化碳还原中间体的结构,发现中间体吸附能与其吸附构型具有显著的相关关系。

关键词: 二氧化碳, 电化学, 催化剂, DFT计算, 机器学习

Abstract:

Electrocatalytic reduction of carbon dioxide to high value-added chemical products has provided a new route to alleviate greenhouse effect and other global problems, attracting intensive attention from both industry and academia. However, it still remains a great challenge to develop electrocatalysts with high performance for practical applications. As one of the major products from carbon dioxide electroreduction, formate is of key importance due to its stability, portability, and high volumetric energy density. In this work, we systematically studied the performance of electrochemical reduction of carbon dioxide to formate on various dual-metal-site catalysts using machine learning in conjunction with density functional theory (DFT) calculations. It was determined that the atomic number, electronegativity and ionization energy of metal atoms in the reaction center are major factors influencing the adsorption of formate intermediates over dual-metal-site catalysts. Based on these features, we predicted the adsorption free energy change for the electroreduction of carbon dioxide to formate and its main competitive reaction, hydrogen evolution reaction. 29 out of 105 dual-metal-site catalysts were identified as potential catalysts for formate production from carbon dioxide electroreduction. A similar method was used to predict the structure of the carbon dioxide reduction intermediates on the surface of 105 dual-metal-site catalysts, and it was found that the adsorption energy of the intermediates has a significant correlation with their adsorption configuration.

Key words: carbon dioxide, electrochemistry, catalyst, DFT calculation, machine learning

中图分类号: 

  • O 646.51

图1

石墨烯-N6-M1-M2双金属中心催化剂模型[灰色球为碳原子,淡蓝色球为氮原子,深蓝色球为金属原子(M1和M2)]"

图2

CO2ER和HER关键反应中间体的不同吸附构型[灰色球为碳原子,淡蓝色球为氮原子,白色球为氢原子,红色球为氧原子,深蓝色球为金属原子(M1和M2)]"

表1

随机抽取的23种DMSCs的ΔG*H、ΔG*HCOO和*HCOO吸附构型"

序号M1M2*HCOO吸附构型ΔG*H/eVΔG*HCOO/eV
1CuNiM1-*HCOO-M2双配位型0.430.89
2CoPdM-*HCOO单配位扭曲型-0.760.26
3RhIrM-*HCOO单配位扭曲型-0.500.51
4RhPtM-*HCOO单配位扭曲型-0.610.54
5IrCrM-*HCOO单配位型-0.32-0.43
6IrMnM-*HCOO单配位型-0.320.00
7ZnFeM1-*HCOO-M2双配位型-0.45-0.85
8RuFeM1-*HCOO-M2双配位型-0.050.12
9CrCoM1-*HCOO-M2双配位型-0.29-0.48
10MnAgM-*HCOO单配位型-0.66-0.62
11AgCoM1-*HCOO-M2双配位型0.080.27
12CrPdM-*HCOO单配位型-0.43-0.77
13NiPtM-*HCOO单配位扭曲型-0.341.31
14AgAuM-*HCOO单配位型-0.240.92
15MnPtM-*HCOO单配位型-0.61-0.53
16ZnNiM1-*HCOO-M2双配位型-0.31-0.25
17RuCuM1-*HCOO-M2双配位型-0.68-0.13
18CuAgM1-*HCOO-M2双配位型0.130.13
19NiCoM1-*HCOO-M2双配位型-0.590.53
20RuCoM1-*HCOO-M2双配位型-0.240.34
21CrMnM1-*HCOO-M2双配位型-0.18-1.21
22RhAgM-*HCOO单配位型-0.020.96
23PdZnM-*HCOO单配位型-0.72-0.38

表2

机器学习的完整52个特征"

特征定义特征定义
Z1Z2M1和M2的原子序数(Z1+Z2)/2平均原子序数
N1N2M1和M2的价电子数(N1+N2)/2平均价电子数
PE1PE2M1和M2的电负性[34](PE1+PE2)/2平均电负性
IE1IE2M1和M2的第一电离能[34](IE1+IE2)/2平均第一电离能
EA1EA2M1和M2的电子亲和能[35](EA1+EA2)/2平均电子亲和能
Nd1Nd2M1和M2的d电子数(Nd1+Nd2)/2平均d电子数
WF1WF2M1和M2的功函数[34](WF1+WF2)/2平均功函数
r1r2M1和M2的原子半径[34](r1+r2)/2平均原子半径
R1R2M1和M2的范德华半径[36](R1+R2)/2平均范德华半径
IE1/Nd1IE2/Nd2第一电离能除以d电子数[(Z1+Z2)/2]2平均原子序数的平方值
EA1/Nd1EA2/Nd2电子亲和能除以d电子数[(N1+N2)/2]2平均价电子数的平方值
PE1/Nd1PE2/Nd2电负性除以d电子数[(PE1+PE2)/2]2平均电负性的平方值
PE1×Nd1PE2×Nd2电负性与d电子数之积[(IE1+IE2)/2]2平均第一电离能的平方值
PE1+PE2PE1-PE2电负性之和、电负性之差[(EA1+EA2)/2]2平均电子亲和能的平方值
r1+r2原子半径之和[(Nd1+Nd2)/2]2平均d电子数的平方值
R1+R2范德华半径之和[(WF1+WF2)/2]2平均功函数的平方值
WF1/Nd1WF2/Nd2功函数除以d电子数[(r1+r2)/2]2平均原子半径的平方值
WF1+WF2、WF1-WF2功函数之和、功函数之差[(R1+R2)/2]2平均范德华半径的平方值

图3

ΔG*HCOO的特征筛选研究"

图4

8种算法对ΔG*HCOO预测结果的对比"

图5

8种算法预测*HCOO吸附构型的准确率"

图6

ΔG*HCOO的7个特征在训练集和全数据集下的归一化相对数值的出现频率分布情况"

图7

机器学习对ΔG*HCOO的预测结果"

图8

机器学习对ΔG*H的预测结果"

图9

机器学习对ΔG*HCOO-ΔG*H的预测结果"

图10

机器学习对*HCOO吸附构型的预测结果"

1 Herron J A, Kim J, Upadhye A A, et al. A general framework for the assessment of solar fuel technologies[J]. Energy & Environmental Science, 2015, 8(1): 126-157.
2 Davis S J, Caldeira K, Matthews H D. Future CO2 emissions and climate change from existing energy infrastructure[J]. Science, 2010, 329(5997): 1330-1333.
3 Wu J H, Huang Y, Ye W, et al. CO2 reduction: from the electrochemical to photochemical approach[J]. Advanced Science, 2017, 4(11): 1700194.
4 Bonetto R, Crisanti F, Sartorel A. Carbon dioxide reduction mediated by iron catalysts: mechanism and intermediates that guide selectivity[J]. ACS Omega, 2020, 5(34): 21309-21319.
5 Li F W, Chen L, Xue M Q, et al. Towards a better Sn: efficient electrocatalytic reduction of CO2 to formate by Sn/SnS2 derived from SnS2 nanosheets[J]. Nano Energy, 2017, 31: 270-277.
6 Chen C, Khosrowabadi Kotyk J F, Sheehan S W. Progress toward commercial application of electrochemical carbon dioxide reduction[J]. Chem, 2018, 4(11): 2571-2586.
7 Lu Q, Jiao F. Electrochemical CO2 reduction: electrocatalyst, reaction mechanism, and process engineering[J]. Nano Energy, 2016, 29: 439-456.
8 Hui S R, Shaigan N M, Neburchilov V, et al. Three-dimensional cathodes for electrochemical reduction of CO2: from macro-to nano-engineering[J]. Nanomaterials, 2020, 10(9): 1884.
9 Gao W L, Liang S Y, Wang R J, et al. Industrial carbon dioxide capture and utilization: state of the art and future challenges[J]. Chemical Society Reviews, 2020, 49(23): 8584-8686.
10 Yang M R, Li H J, Luo N D, et al. Electro-chemical reduction of carbon dioxide into ethylene: catalyst, conditions and mechanism[J]. Progress in Chemistry, 2019, 31(2/3): 245-257.
11 Chen S H, Su Y Q, Deng P L, et al. Highly selective carbon dioxide electroreduction on structure-evolved copper perovskite oxide toward methane production[J]. ACS Catalysis, 2020, 10(8): 4640-4646.
12 Yu J L, Liu H Y, Song S Q, et al. Electrochemical reduction of carbon dioxide at nanostructured SnO2/carbon aerogels: the effect of tin oxide content on the catalytic activity and formate selectivity[J]. Applied Catalysis A: General, 2017, 545: 159-166.
13 Agarwal A S, Zhai Y M, Hill D, et al. The electrochemical reduction of carbon dioxide to formate/formic acid: engineering and economic feasibility[J]. ChemSusChem, 2011, 4(9): 1301-1310.
14 Loges B, Boddien A, Gärtner F, et al. Catalytic generation of hydrogen from formic acid and its derivatives: useful hydrogen storage materials[J]. Topics in Catalysis, 2010, 53(13/14): 902-914.
15 Li J, Jiao J Q, Zhang H C, et al. Two-dimensional SnO2 nanosheets for efficient carbon dioxide electroreduction to formate[J]. ACS Sustainable Chemistry & Engineering, 2020, 8(12): 4975-4982.
16 Mou K W, Chen Z P, Yao S Y, et al. Enhanced electrochemical reduction of carbon dioxide to formate with in situ grown indium-based catalysts in an aqueous electrolyte[J]. Electrochimica Acta, 2018, 289: 65-71.
17 Fu Y S, Li Y N, Zhang X, et al. Novel hierarchical SnO2 microsphere catalyst coated on gas diffusion electrode for enhancing energy efficiency of CO2 reduction to formate fuel[J]. Applied Energy, 2016, 175: 536-544.
18 Chen Y H, Kanan M W. Tin oxide dependence of the CO2 reduction efficiency on tin electrodes and enhanced activity for tin/tin oxide thin-film catalysts[J]. Journal of the American Chemical Society, 2012, 134(4): 1986-1989.
19 Pander J E III, Baruch M F, Bocarsly A B. Probing the mechanism of aqueous CO2 reduction on post-transition-metal electrodes using ATR-IR spectroelectrochemistry[J]. ACS Catalysis, 2016, 6(11): 7824-7833.
20 Zhu X R, Yan J X, Gu M, et al. Activity origin and design principles for oxygen reduction on dual-metal-site catalysts: a combined density functional theory and machine learning study[J]. The Journal of Physical Chemistry Letters, 2019, 10(24): 7760-7766.
21 He Q, Yu B, Li Z H, et al. Density functional theory for battery materials[J]. Energy & Environmental Materials, 2019, 2(4): 264-279.
22 Li M R, Garg S, Chang X X, et al. Toward excellence of transition metal-based catalysts for CO2 electrochemical reduction: an overview of strategies and rationales[J]. Small Methods, 2020, 4(7): 2000033.
23 Yang Z, Gao W, Jiang Q. A machine learning scheme for the catalytic activity of alloys with intrinsic descriptors[J]. Journal of Materials Chemistry A, 2020, 8(34): 17507-17515.
24 Chen Y, Huang Y, Cheng T, et al. Identifying active sites for CO2 reduction on dealloyed gold surfaces by combining machine learning with multiscale simulations[J]. Journal of the American Chemical Society, 2019, 141(29): 11651-11657.
25 Zhong M, Tran K, Min Y, et al. Accelerated discovery of CO2 electrocatalysts using active machine learning[J]. Nature, 2020, 581(7807): 178-183.
26 Ma X, Li Z, Achenie L E, et al. Machine-learning-augmented chemisorption model for CO2 electroreduction catalyst screening[J]. The Journal of Physical Chemistry Letters, 2015, 6(18): 3528-3533.
27 Mayer F D, Hosseini-Benhangi P, Sánchez-Sánchez C M, et al. Scanning electrochemical microscopy screening of CO2 electroreduction activities and product selectivities of catalyst arrays[J]. Communications Chemistry, 2020, 3: 155.
28 Wu D H, Zhang J Y, Cheng M J, et al. Machine learning investigation of supplementary adsorbate influence on copper for enhanced electrochemical CO2 reduction performance[J]. The Journal of Physical Chemistry C, 2021, 125(28): 15363-15372.
29 Zhang H C, Goddard W A, Lu Q, et al. The importance of grand-canonical quantum mechanical methods to describe the effect of electrode potential on the stability of intermediates involved in both electrochemical CO2 reduction and hydrogen evolution[J]. Physical Chemistry Chemical Physics, 2018, 20(4): 2549-2557.
30 Ma W C, Xie S J, Zhang X G, et al. Promoting electrocatalytic CO2 reduction to formate via sulfur-boosting water activation on indium surfaces[J]. Nature Communications, 2019, 10: 892.
31 Hastie T, Tibshirani R, Friedman J H, et al. The elements of statistical learning: data mining, inference, and prediction[J]. The Mathematical Intelligencer, 2004, 27(2): 83-85.
32 Friedman J H. Greedy function approximation: a gradient boosting machine[J]. The Annals of Statistics, 2001, 29(5): 1189-1232.
33 Kim J, Jung H, Jung S M, et al. Tailoring binding abilities by incorporating oxophilic transition metals on 3D nanostructured Ni arrays for accelerated alkaline hydrogen evolution reaction[J]. Journal of the American Chemical Society, 2021, 143(3): 1399-1408.
34 Speight J G. Lange's Handbook of Chemistry[M]. New York: McGraw-Hill, 2005: 1.132-1.156.
35 喻典, 梁国明. 元素电子亲和势的密度泛函理论计算[J]. 重庆师范大学学报(自然科学版), 2005, 22(1): 39-42.
Yu D, Liang G M. A study of electron affinities of the elements by density functional theory[J]. Journal of Chongqing Teachers College (Natural Science Edition), 2005, 22(1): 39-42.
36 Alvarez S. A cartography of the van der Waals territories[J]. Dalton Transactions, 2013, 42(24): 8617.
[1] 戴晓业, 安青松, 许云婷, 史琳. 废弃制冷剂降解方法研究现状及思考[J]. 化工学报, 2021, 72(S1): 1-6.
[2] 李敏霞, 詹浩淼, 王派, 刘雪涛, 李昱翰, 马一太. 一种带引射器和经济器的CO2跨临界制冷系统[J]. 化工学报, 2021, 72(S1): 146-152.
[3] 李腾飞, 缪赟, 杨柳, 王龙耀, 朱铧丞. 微波强化Y型分子筛离子交换技术[J]. 化工学报, 2021, 72(S1): 406-412.
[4] 高帅涛, 刘雪珂, 张丽, 刘芬, 余江, 商剑锋, 欧天雄, 周政, 陈平文. Aspen Plus模拟高浓度H2S/CO2酸性气的选择性分离[J]. 化工学报, 2021, 72(S1): 413-420.
[5] 吴俊晔, 葛天舒, 吴宣楠, 代彦军, 王如竹. 基于吸附剂/木浆纤维纸耦合材料的空气净化[J]. 化工学报, 2021, 72(S1): 520-529.
[6] 梁家豪, 张国强, 高源, 尹娇, 郑华艳, 李忠. 介孔构建对CuY甲醇氧化羰基化反应活性的影响[J]. 化工学报, 2021, 72(9): 4685-4697.
[7] 颜建国, 郑书闽, 郭鹏程, 张博, 毛振凯. 基于GA-BP神经网络的超临界CO2传热特性预测研究[J]. 化工学报, 2021, 72(9): 4649-4657.
[8] 王欢, 符方宝, 李琼, 席跃宾, 杨东杰. 木质素碳纳米材料制备及在催化中的应用研究进展[J]. 化工学报, 2021, 72(9): 4445-4457.
[9] 谢晶, 舒歌平, 杨葛灵, 高山松, 王洪学, 卢晗锋, 陈银飞. Mo修饰的钼铁复合催化剂及其煤直接液化催化性能[J]. 化工学报, 2021, 72(9): 4675-4684.
[10] 单良, 尹荣强, 王慧, 费传军, 周清清, 徐杰, 王志强, 徐涛, 陈建军, 李俊华. VMoTi/玻纤复合催化滤布制备及其除尘协同脱硝性能研究[J]. 化工学报, 2021, 72(9): 4892-4899.
[11] 从少领, 赵捷, 杨玉飞, 吴长清, 贺凡, 袁华, 汪晓芹, 熊善新, 吴燕, 周安宁. 煤基聚苯胺制掺N碳微纳米管的实验研究[J]. 化工学报, 2021, 72(9): 4950-4960.
[12] 梁旭鸣, 沈永超, 卫东, 郭倩, 高志. 基于直流内阻和交流阻抗特性的铝空气电池输出特性分析[J]. 化工学报, 2021, 72(8): 4361-4370.
[13] 李宇明, 刘梓烨, 张启扬, 王雅君, 崔国庆, 姜桂元, 贺德华. 氮掺杂碳材料的制备及其在催化领域中的应用[J]. 化工学报, 2021, 72(8): 3919-3932.
[14] 冯晓博, 刘天龙, 赵小燕, 曹景沛. 合成气与二甲醚为原料直接制乙醇催化反应研究进展[J]. 化工学报, 2021, 72(8): 3958-3967.
[15] 赵岩, 李秀萍, 赵荣祥. 苯酚型低共熔溶剂中硫酸钛作为催化剂高效氧化脱硫[J]. 化工学报, 2021, 72(8): 4391-4400.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 黄福明,张国伟,胡春圃,应圣康. 硅溶胶/苯乙烯-丙烯酸酯共聚物无机-有机杂化水分散液的制备和表征[J]. CIESC Journal, 2005, 13(6): 816 -823 .
[2] 姬忠礼,焦海青,陈鸿海. 陶瓷过滤管表面粉尘层清除过程的图像分析[J]. CIESC Journal, 2005, 13(2): 178 -183 .
[3] 孙宝芝,淮秀兰,姜任秋,刘登瀛. 声空强化渗透脱水过程质扩散研究[J]. CIESC Journal, 2005, 13(1): 13 -17 .
[4] 王微微. 油气两相流空隙率测量[J]. CIESC Journal, 2007, 15(3): 339 -344 .
[5] 张瑛, 窦涛, 鲍晓军, 李玉平, 李晓峰. 基于固相转化机理合成五元环沸石的新技术[J]. CIESC Journal, 2003, 11(6): 656 -659 .
[6] 刘先桥, 官月平, 邢建民, 马志亚, 刘会洲. 带环氧基的超顺磁性高分子微球的制备及其性能表征[J]. CIESC Journal, 2003, 11(6): 731 -735 .
[7] 祝贵兵, 彭永臻, 王淑莹, 左金龙, 王亚宜, 郭建华. 分段进水生物脱氮工艺稳态模型的开发与试验评价[J]. CIESC Journal, 2007, 15(3): 411 -417 .
[8] 司徒粤, 胡剑峰, 黄洪, 傅和青, 曾汉维, 陈焕钦. 新型环氧大豆油增韧酚醛树脂的合成与性质[J]. CIESC Journal, 2007, 15(3): 418 -423 .
[9] 潘海华, 李啸风, 李浩然, 刘迪霞, 韩世钧. 双亲分子在油水界面的行为研究[J]. CIESC Journal, 2003, 11(4): 446 -451 .
[10] 柯明, 汪燮卿, 张凤美. 磷改性ZSM-5分子筛物化性质和裂解制乙烯性能的研究[J]. CIESC Journal, 2003, 11(6): 671 -676 .