Ion Transport in Membranes with Sub-Nanometer Pores: Material Design and Selectivity Mechanisms

Date of Award

Spring 2022

Document Type


Degree Name

Doctor of Philosophy (PhD)


Chemical and Environmental Engineering (ENAS)

First Advisor

Elimelech, Menachem


As of 2017, over 800 million people around the world still lacked the access to safe drinking water. With over 97% of the earth’s water stored in the ocean, reverse osmosis (RO) seawater desalination has been proposed as a promising solution to boost the water supply and alleviate the water shortage. Polyamide (PA) thin-film composite (TFC) membranes are the state-of-the-art membranes for RO separation and a typical PA TFC membrane can reject over 99% of the sodium chloride (NaCl) in the feed stream. Despite their exceptional water-salt separation capability, current PA membranes have two major limitations: 1) they are highly sensitive to oxidants, which restrains the use of chlorine cleaning in controlling the membrane biofouling, and 2) their ability to differentiate the transport of different solutes is inadequate, which limits the further application of membrane separation in new areas such as resource recovery (e.g., recovering lithium from the seawater). In this dissertation we explore advancing the current membrane desalination technology in these two aspects. Specifically, robust membranes with high water-solute selectivity are developed by both enhancing the chemical stability of current PA TFC membranes via inorganic surface coating, and by creating advanced membranes with high selectivity directly with ceramic materials. Additionally, we study mechanisms underlying the solute transport and selectivity through subnanometer pores with the emphasis on the impact of the ion-ion, ion-pore, and ion-water interactions. Understanding these interactions is crucial for the fabrication of membranes with the desired selectivity.First, we explore the feasibility of creating a uniform titanium oxide (TiO2) protection layer on top of the PA surface using the atomic layer deposition (ALD) technique. Deposition conditions such as precursor dosage and exposure were optimized for both PA RO and nanofiltration substrates, with the purpose of enhancing the coating uniformity. Notably, with the optimized deposition conditions, we demonstrated the creation of a uniform TiO2 coating layer with a thickness of a few atoms, even on the ridge-and-valley structure of PA RO membranes. This uniform TiO2 coating layer created enhanced chlorine resistance of the PA TFC membranes. Inspired by the observed water permeation with the PA membranes coated with a thick (> 5 nm) TiO2 ALD layer, we then discuss the feasibility of creating dense ceramic membranes with subnanometer pores using the ALD technique. Specifically, we proved the existence of subnanometer voids within the ALD deposited TiO2 films (4 Å < d < 12 Å, with two distinct peaks at 5.5 and 6.5 Å) and suggested that these voids are created by the hindered diffusion of ALD precursors (tetrakistitanium; d = 7 Å). Further membrane conductance measurement revealed a permeation of sodium fluoride more than eight times faster than other sodium halides through the subnanometer TiO2 voids. Intrigued by the observed selective fluoride transport, we then systematically discuss the mechanisms underlying the ion transport through subnanometer pores, with the emphasis on the impact of ion-water, ion-ion, and ion-pore interactions. Through comparing the energy barriers for the transport of individual ions and the energy barrier for salt transport, we suggest that the cation and anion traverse the membrane pores individually during salt transport, with each ion experiencing a distinct energy barrier. Notably, we observed a higher energy barrier for counterion permeation compared to coions even if the counterion has a lower hydration energy. This observation challenges the accepted view on the dominance of ion dehydration at the pore entrance in determining the energy barrier for ion transport, and implies the significant contribution from the hindered intrapore diffusion due to ion-pore attraction. Lastly, we investigate the further tuning of the pore structure within the ALD-deposited ceramic films; through incorporating the carbon species (i.e., methoxyl groups) within the inorganic films (i.e., aluminum oxide) and burning them out after deposition, we demonstrated the creation of additional pore structures within the inorganic films. The selectivity of the fabricated membranes over the transport of salts of different valences was studied and a ten times faster sodium chloride permeation was observed compared to the flux of sodium sulfate or calcium chloride. Analysis of the dependence of monovalent/divalent salt selectivity on the membrane pore size suggested that the hydration/dehydration status of the ions, rather than the extent of ion dehydration, plays the dominant role in differentiating the transport of ions with different hydrated radius (e.g., monovalent and divalent salts). Overall, this dissertation introduces the development of the TiO2-protected PA membranes and the ceramic membranes with tunable subnanometer pores and discusses how ion pairing and ion dehydration impact the salt permeation in a confined environment. Insights provided in this work should guide the development of membrane processes with enhanced sustainability and help extend the application of membrane technologies in areas such as resource recovery.

This document is currently not available here.