Yi Zhao , Hao Li , Yu Chen , Dingwen Tong , Yong Bao , Xiaofeng Qiu , Haizhen Sun , Hao Yang
{"title":"High-flexibility droplet generation via centrifugal capillary array platforms with machine learning-driven dimensional precision","authors":"Yi Zhao , Hao Li , Yu Chen , Dingwen Tong , Yong Bao , Xiaofeng Qiu , Haizhen Sun , Hao Yang","doi":"10.1016/j.colsurfa.2025.137466","DOIUrl":null,"url":null,"abstract":"<div><div>Droplet microfluidics has emerged as a pivotal technology in biomedical, materials, and environmental engineering due to its precise microscale manipulation capabilities. However, conventional methods relying on microfabricated channels and external pumps face limitations in dynamic adjustability, scalability, and practical applicability. To address these challenges, this study introduces a centrifugal capillary array platform (CCAP) for high-throughput, pump-free droplet generation. The system leverages centrifugal force-driven fluid dynamics and machine learning-enabled detection to achieve precise control over monodisperse droplet formation. Through integration of multifaceted parameter optimization—encompassing capillary geometry, rotational kinematics, and interfacial properties—the platform exhibits robust operational stability and tunable droplet dimensions, thereby offering distinct advantages over conventional microfluidic approaches constrained by fixed geometries and pump-induced instabilities. A deep learning-based detection framework, enhanced with Hough circle optimization, enables high-precision droplet characterization with sub-1 % measurement accuracy, establishing a foundation for closed-loop process control. The CCAP’s versatility is further validated through the successful construction of biphasic W/O and O/W systems and the synthesis of functionalized magnetic porous trimethylolpropane triacrylate (TMPTA) particles. These particles exhibit exceptional catalytic performance in degrading organic pollutants such as methylene blue, achieving rapid pollutant removal while maintaining robust recyclability. By eliminating the need for microfabrication and external pumps inherent to conventional methods, the CCAP platform delivers a scalable, intelligent, and cost-effective solution for applications including precision material synthesis; environmental remediation; and high-throughput biochemical analysis.</div></div>","PeriodicalId":278,"journal":{"name":"Colloids and Surfaces A: Physicochemical and Engineering Aspects","volume":"724 ","pages":"Article 137466"},"PeriodicalIF":4.9000,"publicationDate":"2025-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Colloids and Surfaces A: Physicochemical and Engineering Aspects","FirstCategoryId":"92","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S092777572501369X","RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}
引用次数: 0
Abstract
Droplet microfluidics has emerged as a pivotal technology in biomedical, materials, and environmental engineering due to its precise microscale manipulation capabilities. However, conventional methods relying on microfabricated channels and external pumps face limitations in dynamic adjustability, scalability, and practical applicability. To address these challenges, this study introduces a centrifugal capillary array platform (CCAP) for high-throughput, pump-free droplet generation. The system leverages centrifugal force-driven fluid dynamics and machine learning-enabled detection to achieve precise control over monodisperse droplet formation. Through integration of multifaceted parameter optimization—encompassing capillary geometry, rotational kinematics, and interfacial properties—the platform exhibits robust operational stability and tunable droplet dimensions, thereby offering distinct advantages over conventional microfluidic approaches constrained by fixed geometries and pump-induced instabilities. A deep learning-based detection framework, enhanced with Hough circle optimization, enables high-precision droplet characterization with sub-1 % measurement accuracy, establishing a foundation for closed-loop process control. The CCAP’s versatility is further validated through the successful construction of biphasic W/O and O/W systems and the synthesis of functionalized magnetic porous trimethylolpropane triacrylate (TMPTA) particles. These particles exhibit exceptional catalytic performance in degrading organic pollutants such as methylene blue, achieving rapid pollutant removal while maintaining robust recyclability. By eliminating the need for microfabrication and external pumps inherent to conventional methods, the CCAP platform delivers a scalable, intelligent, and cost-effective solution for applications including precision material synthesis; environmental remediation; and high-throughput biochemical analysis.
期刊介绍:
Colloids and Surfaces A: Physicochemical and Engineering Aspects is an international journal devoted to the science underlying applications of colloids and interfacial phenomena.
The journal aims at publishing high quality research papers featuring new materials or new insights into the role of colloid and interface science in (for example) food, energy, minerals processing, pharmaceuticals or the environment.