Janhavi Nistane
Advisor: Prof. Rampi Ramprasad
will propose a doctoral thesis entitled,
Informatics-driven sustainable polymer membrane design for binary solvent separations
On
Monday, Sept. 23, 2024
10am - 12pm
MRDC Room 3515
or virtually via Teams:
Meeting ID: 288 224 298 458
Committee:
Prof. Rampi Ramprasad- School of Materials Science and Engineering (advisor)
Prof. Seung Soon Jang- School of Materials Science and Engineering (co-advisor)
Prof. Ryan Lively- School of Chemical and Biomolecular Engineering
Prof. Aaron Stebner- School of Mechanical Engineering and Materials Science and Engineering
Prof. Guoxiang (Emma) Hu - School of Materials Science and Engineering
Abstract: Refineries consume vast amounts of energy to separate crude oil-based organic sol- contributing to 10-15 % of the total energy consumption in the United States. Alternatively, polymer membranes achieve the same separations using only one-tenth of the energy. Despite their potential, commercial adoption of membrane-based separations remains limited. This limitation stems from the difficulty of identifying membranes that offer both high permeability (enabling rapid transport) and high selectivity (allowing for effective molecular discrimination), as these properties are often inversely related. This inverse relationship is typically illustrated using permeability-selectivity trade-off plots, or Robeson plots, which are essential in the gas membrane industry for identifying optimal membranes that exceed established performance bounds. While trade-off plots are well-established for gas separations, a similar benchmark for solvent separations is lacking, impeding the discovery of effective polymer membranes. Additionally, the environmental concerns associated with the toxicity of commercially used halogenated membranes raise the urgent need to find sustainable alternatives. However, traditional methods of experimentation and simulations are insufficient to navigate the vast polymer landscape and discover suitable and sustainable membranes. To address these challenges, we present a data-driven pipeline to identify sustainable polymer membranes for binary organic solvent separations that achieve desired target properties. This methodology enables the construction of the largest known separation-based solvent trade-off plots for 13,000 polymers. Furthermore, in our pursuit of sustainable alternatives, we have screened over 5 million virtually generated depolymerizable candidates and discovered potential replacements for halogenated membranes. This contribution is expected to significantly accelerate the discovery of polymer membranes for solvent separations.