Simulation of Reverse Osmosis Membrane Compaction Using Material Point Method (MPM)

Sreejith N. A., Hariswaran Sitaraman, Marc Day, Yara Suleiman, Sina Shabazmohamadi, Jishan Wu, Eric Hoek

Research output: NRELPoster


Access to fresh drinking water in the future requires effective management of industrial wastewater and water purification from available resources. In this study, we present the simulation methodology and the analysis of a Reverse Osmosis (RO) membrane under various pressure conditions using the Material Point Method (MPM). In contrast to other numerical methods, MPM solves the continuum governing equations on material points in a Lagrangian framework. The method does not require a grid connecting the material points hence making it suitable to simulate large deformations during membrane compaction. The time integration is carried out using the explicit Euler method, while the spatial discretization is performed using linear or cubic-spline shape functions. A series of planar images containing detailed pore structures obtained from X-ray tomography experiments is converted to a three-dimensional collection of material points to simulate the membrane. Compressive loads are applied to the top layer of the membrane to simulate the experiments. The membrane deformation and pore size distribution before and after load application are reported and compared with the experimental measurements. The presentation discusses the numerical methods used, the performance of the solver on high-performance computing machines, and the results of membrane compaction in detail.
Original languageAmerican English
StatePublished - 2023

Publication series

NamePresented at the National Alliance for Water Innovation (NAWI) Quarterly Review, 28 February - 1 March 2023

NREL Publication Number

  • NREL/PO-2C00-85460


  • high pressure reverse osmosis
  • Material Point Method
  • membrane compaction


Dive into the research topics of 'Simulation of Reverse Osmosis Membrane Compaction Using Material Point Method (MPM)'. Together they form a unique fingerprint.

Cite this