4.7 Article

Optimizing pressure retarded osmosis spacer geometries: An experimental and CFD modeling study

期刊

JOURNAL OF MEMBRANE SCIENCE
卷 647, 期 -, 页码 -

出版社

ELSEVIER
DOI: 10.1016/j.memsci.2022.120284

关键词

Computational fluid dynamics; Membrane spacer; Geometry optimization; Single-objective genetic algorithms; Concentration polarization

资金

  1. Bureau of Reclamation (US Department of Interior) [R19AC00100]
  2. GEM Foundation
  3. Alfred P. Sloan Foundation
  4. USF Presidential Fellowship
  5. Florida Education Fund

向作者/读者索取更多资源

Existing desalination plants face challenges of high energy costs and environmental impacts. Pressure retarded osmosis (PRO) technology can mitigate these issues by capturing the potential energy in the salinity gradient. However, PRO faces the challenge of concentration polarization (CP), which can be reduced using membrane spacers but they also increase the pressure drop along the feed channel.
Existing desalination plants face the challenges of high energy costs and environmental impacts from brine disposal. Pressure retarded osmosis (PRO) is a membrane-based technology that could mitigate these issues by capturing the potential energy in the salinity gradient between brine and dilute (e.g., freshwater/wastewater) solutions. Currently, a major challenge facing PRO is concentration polarization (CP), which is the reduction of the salinity gradient caused by the buildup of solutes along both the brine side of the membrane surface (external) and within the membrane support structure (internal). CP can be reduced using membrane spacers, but they have the drawback of increasing the pressure drop along the length of the feed channel. Therefore, it is important to design a membrane spacer that can minimize the effects of both CP and the internal pressure drop.In this study, we simulated the effect that membrane spacer geometries have on CP and the internal pressure drop via computational fluid dynamics (CFD) using the open-source software, OpenFOAM. Simulations were validated using experimental results for five physical spacer geometries on a bench-scale PRO system. The highest performing physical spacer geometry was a 47 mil (1.41 mm) thick parallel oriented spacer (47P), which was then improved upon through integrated CFD-based optimization using the open-source optimization toolkit, DAKOTA. Single-objective genetic algorithms were used to iteratively modify the spacer geometry by increasing the linearity of the thinner cross-strand and decreasing the linearity of the thicker parallel strand, which created a more hexagonally shaped cross-section. This modification was able to increase overall system performance, which was represented by the ratio of the mass flux density of the feed permeate through the membrane to the pressure drop across the length of the feed channel, by 16.3%. This suggests that a membrane spacer with a parallel hexagonal configuration allows the fluid to flow with minimal internal pressure drop while still creating the necessary back-mixing from the membrane surface to the bulk fluid needed to promote mass transfer and reduce the impacts of CP. Implementing these improvements in spacer design would cause the feed pump to consume less energy, which would increase the net energy generation potential and overall sustainability of PRO systems.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据