Xiangyu Song , Shuoyang Wang , Fan Wang , Yingwu Liu , Zongliang Zuo , Siyi luo , Dong Chen , Fangchao Zhao
{"title":"机器学习辅助催化剂合成和硼氢化钠催化水解制氢","authors":"Xiangyu Song , Shuoyang Wang , Fan Wang , Yingwu Liu , Zongliang Zuo , Siyi luo , Dong Chen , Fangchao Zhao","doi":"10.1016/j.ijhydene.2025.04.286","DOIUrl":null,"url":null,"abstract":"<div><div>As a clean and renewable energy source, hydrogen plays a crucial role in achieving sustainable energy goals. However, the high cost of hydrogen production remains a major challenge. Therefore, research into hydrogen production technology is of significant importance. Sodium borohydride (NaBH<sub>4</sub>), due to its potential to generate hydrogen through hydrolysis, is widely recognized as an efficient hydrogen storage material. In this study, the Co–P–B/ZIF-67 catalyst was successfully prepared using a chemical reduction method, and for the first time, machine learning technology was applied to predict and optimize the NaBH<sub>4</sub> hydrolysis hydrogen production process. The findings reveal that the hydrogen production yield of Co–P–B/ZIF-67 is 8.86 times that of traditional Co–P–B and 0.065 times that of ZIF-67, demonstrating a notable catalytic advantage.Additionally, parameters such as catalyst dosage and reactant concentration were found to significantly impact the time required to reach saturated hydrogen production. Through machine learning optimization, it was discovered that increasing the Na<sup>+</sup> concentration can substantially enhance hydrogen production. However, ZIF-67 may occupy some of the active sites of Co–P–B, thereby weakening the catalytic efficiency to some extent. Among the models tested, the random forest algorithm performed the best, with an R<sup>2</sup> value ranging from 0.956 to 0.995, and the number of outliers was reduced from 9 to 0 after optimization, greatly improving the prediction accuracy and stability. Furthermore, machine learning was employed to analyze the mechanism by which different reaction conditions influence hydrogen production, providing new theoretical foundations and technical support for catalyst design and optimization.This interdisciplinary approach not only showcases the immense potential of machine learning in predicting complex chemical reactions but also offers new insights for practical applications in hydrogen energy production.</div></div>","PeriodicalId":337,"journal":{"name":"International Journal of Hydrogen Energy","volume":"129 ","pages":"Pages 130-149"},"PeriodicalIF":8.1000,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Machine learning-assisted catalyst synthesis and hydrogen production via catalytic hydrolysis of sodium borohydride\",\"authors\":\"Xiangyu Song , Shuoyang Wang , Fan Wang , Yingwu Liu , Zongliang Zuo , Siyi luo , Dong Chen , Fangchao Zhao\",\"doi\":\"10.1016/j.ijhydene.2025.04.286\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>As a clean and renewable energy source, hydrogen plays a crucial role in achieving sustainable energy goals. However, the high cost of hydrogen production remains a major challenge. Therefore, research into hydrogen production technology is of significant importance. Sodium borohydride (NaBH<sub>4</sub>), due to its potential to generate hydrogen through hydrolysis, is widely recognized as an efficient hydrogen storage material. In this study, the Co–P–B/ZIF-67 catalyst was successfully prepared using a chemical reduction method, and for the first time, machine learning technology was applied to predict and optimize the NaBH<sub>4</sub> hydrolysis hydrogen production process. The findings reveal that the hydrogen production yield of Co–P–B/ZIF-67 is 8.86 times that of traditional Co–P–B and 0.065 times that of ZIF-67, demonstrating a notable catalytic advantage.Additionally, parameters such as catalyst dosage and reactant concentration were found to significantly impact the time required to reach saturated hydrogen production. Through machine learning optimization, it was discovered that increasing the Na<sup>+</sup> concentration can substantially enhance hydrogen production. However, ZIF-67 may occupy some of the active sites of Co–P–B, thereby weakening the catalytic efficiency to some extent. Among the models tested, the random forest algorithm performed the best, with an R<sup>2</sup> value ranging from 0.956 to 0.995, and the number of outliers was reduced from 9 to 0 after optimization, greatly improving the prediction accuracy and stability. Furthermore, machine learning was employed to analyze the mechanism by which different reaction conditions influence hydrogen production, providing new theoretical foundations and technical support for catalyst design and optimization.This interdisciplinary approach not only showcases the immense potential of machine learning in predicting complex chemical reactions but also offers new insights for practical applications in hydrogen energy production.</div></div>\",\"PeriodicalId\":337,\"journal\":{\"name\":\"International Journal of Hydrogen Energy\",\"volume\":\"129 \",\"pages\":\"Pages 130-149\"},\"PeriodicalIF\":8.1000,\"publicationDate\":\"2025-04-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Hydrogen Energy\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0360319925019718\",\"RegionNum\":2,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CHEMISTRY, PHYSICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Hydrogen Energy","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0360319925019718","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}
Machine learning-assisted catalyst synthesis and hydrogen production via catalytic hydrolysis of sodium borohydride
As a clean and renewable energy source, hydrogen plays a crucial role in achieving sustainable energy goals. However, the high cost of hydrogen production remains a major challenge. Therefore, research into hydrogen production technology is of significant importance. Sodium borohydride (NaBH4), due to its potential to generate hydrogen through hydrolysis, is widely recognized as an efficient hydrogen storage material. In this study, the Co–P–B/ZIF-67 catalyst was successfully prepared using a chemical reduction method, and for the first time, machine learning technology was applied to predict and optimize the NaBH4 hydrolysis hydrogen production process. The findings reveal that the hydrogen production yield of Co–P–B/ZIF-67 is 8.86 times that of traditional Co–P–B and 0.065 times that of ZIF-67, demonstrating a notable catalytic advantage.Additionally, parameters such as catalyst dosage and reactant concentration were found to significantly impact the time required to reach saturated hydrogen production. Through machine learning optimization, it was discovered that increasing the Na+ concentration can substantially enhance hydrogen production. However, ZIF-67 may occupy some of the active sites of Co–P–B, thereby weakening the catalytic efficiency to some extent. Among the models tested, the random forest algorithm performed the best, with an R2 value ranging from 0.956 to 0.995, and the number of outliers was reduced from 9 to 0 after optimization, greatly improving the prediction accuracy and stability. Furthermore, machine learning was employed to analyze the mechanism by which different reaction conditions influence hydrogen production, providing new theoretical foundations and technical support for catalyst design and optimization.This interdisciplinary approach not only showcases the immense potential of machine learning in predicting complex chemical reactions but also offers new insights for practical applications in hydrogen energy production.
期刊介绍:
The objective of the International Journal of Hydrogen Energy is to facilitate the exchange of new ideas, technological advancements, and research findings in the field of Hydrogen Energy among scientists and engineers worldwide. This journal showcases original research, both analytical and experimental, covering various aspects of Hydrogen Energy. These include production, storage, transmission, utilization, enabling technologies, environmental impact, economic considerations, and global perspectives on hydrogen and its carriers such as NH3, CH4, alcohols, etc.
The utilization aspect encompasses various methods such as thermochemical (combustion), photochemical, electrochemical (fuel cells), and nuclear conversion of hydrogen, hydrogen isotopes, and hydrogen carriers into thermal, mechanical, and electrical energies. The applications of these energies can be found in transportation (including aerospace), industrial, commercial, and residential sectors.