Machine learning-based approach to identify the optimal design and operation condition of organic solvent nanofiltration (OSN)

Kim, Changsu; You, Chanhee; Ngan, D.; Park, Minseong; Jang, Dongjun; Lee, Sung-Ju; Kim, Jiyong

doi:10.1016/b978-0-323-88506-5.50144-3

Cited by 2 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is hypothesized that the dependence of the separation process on the membrane-and solvent properties might lead to a data space which consists of smaller islands, complicating the generalization of data-driven models. The distinction between data collected from cross-flow versus dead-end processing in studies by Hu et al [6] and Kim et al [45] is an example of this. In their studies, the distinction between these two was captured by one of the descriptors, indicating that they are essentially described by two different models.…”

Section: Data Densitymentioning

confidence: 99%

Organic Solvent Nanofiltration and Data-Driven Approaches

Piccard,

Borges,

Cleuren

et al. 2023

Separations

View full text Add to dashboard Cite

Organic solvent nanofiltration (OSN) is a membrane separation method that has gained much interest due to its promising ability to offer an energy-lean alternative for traditional thermal separation methods. Industrial acceptance, however, is held back by the slow process of membrane screening based on trial and error for each solute-solvent couple to be separated. Such time-consuming screening is necessary due to the absence of predictive models, caused by a lack of fundamental understanding of the complex separation mechanism complicated by the wide variety of solute and solvent properties, and the importance of all mutual solute-solvent-membrane affinities and competing interactions. Recently, data-driven approaches have gained a lot of attention due to their unprecedented predictive power, significantly outperforming traditional mechanistic models. In this review, we give an overview of both mechanistic models and the recent advances in data-driven modeling. In addition to other reviews, we want to emphasize the coherence of all mechanistic models and discuss their relevance in an increasingly data-driven field. We reflect on the use of data in the field of OSN and its compliance with the FAIR principles, and we give an overview of the state of the art of data-driven models in OSN. The review can serve as inspiration for any further modeling activities, both mechanistic and data-driven, in the field.

show abstract

Section: Data Densitymentioning

confidence: 99%

Organic Solvent Nanofiltration and Data-Driven Approaches

Piccard,

Borges,

Cleuren

et al. 2023

Separations

View full text Add to dashboard Cite

show abstract

“…used graph neural networks to identify the most critical solvent parameters affecting the rejection of solutes by polyimide OSN membranes. Other works focused on the optimization of operation conditions, performance parameters such as permeance and rejection, − and solvent–membrane interactions . When validated with experimental data, the applied machine-learning models often showed excellent accuracy and predictive capability. , …”

Section: Introductionmentioning

confidence: 99%

Design of Modified Polymer Membranes Using Machine Learning

Glass,

Schmidt,

Merten

et al. 2024

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Surface modification is an attractive strategy to adjust the properties of polymer membranes. Unfortunately, predictive structure−processing−property relationships between the modification strategies and membrane performance are often unknown. One possibility to tackle this challenge is the application of data-driven methods such as machine learning. In this study, we applied machine learning methods to data sets containing the performance parameters of modified membranes. The resulting machine learning models were used to predict performance parameters, such as the pure water permeability and the zeta potential of membranes modified with new substances. The predictions had low prediction errors, which allowed us to generalize them to similar membrane modifications and processing conditions. Additionally, machine learning methods were able to identify the impact of substance properties and process parameters on the resulting membrane properties. Our results demonstrate that small data sets, as they are common in materials science, can be used as training data for predictive machine learning models. Therefore, machine learning shows great potential as a tool to expedite the development of high-performance membranes while reducing the time and costs associated with the development process at the same time.

show abstract

Machine learning-based approach to identify the optimal design and operation condition of organic solvent nanofiltration (OSN)

Cited by 2 publications

References 3 publications

Organic Solvent Nanofiltration and Data-Driven Approaches

Organic Solvent Nanofiltration and Data-Driven Approaches

Design of Modified Polymer Membranes Using Machine Learning

Contact Info

Product

Resources

About