8/18/2025 Taylor Parks
Written by Taylor Parks
Mechanical science and engineering professor Kyle Smith and mechanical engineering PhD candidate Vu Do recently published a novel desalination technique in Environmental Science & Technology. Their study, “Symmetric Intercalative Desalination of Brackish Water Without Ion-Exchange Membranes,” demonstrates for the first time the successful experimental elimination of ion-exchange membranes (IEMs) from a battery-like desalination device.
“Ion-exchange membranes are significant contributors to the cost of [these] devices, and as such their elimination stands to make brackish water desalination more economical using this battery-based approach,” Smith said.
Where reverse osmosis relies on physically pushing salt particles out of water, battery-based desalination methods leverage the charges present in these particles. When salt dissolves in water, it separates into positively charged sodium ions (i.e., cations) and negatively charged chloride ions (i.e., anions). Battery-based methods use electricity to remove these ions from the water, resulting in a desalinated, or ion-neutral, solution. In some devices, an anion-exchange membrane (AEM), which contains positively charged material, or a cation-exchange membrane (CEM), which contains negatively charged material, helps achieve desalination by allowing the opposite charge to pass through. These membranes also block the diluate and brine from mixing directly. AEMs are often used so that anions can pass to the brine while cations are blocked.
While battery-based methods consume less energy than reverse osmosis for brackish water desalination, they still pose significant manufacturing costs, particularly those that require AEM fabrication. Furthermore, AEMs tend to experience decreased performance (i.e., block less cations) as salt concentration increases. Through experimentation, Do and Smith found that nanofiltration (NF) membranes pose a viable, less expensive alternative to AEMs.
“During experiments we’ve performed over the last few years, we’ve seen that the membranes we use aren’t perfect,” Do said. “This gave us more reason to see what would happen if we completely replaced the membrane.”
Instead of blocking cations like an AEM, an NF membrane allows cations and anions to flow freely, posing an alternative desalination method. The researchers ran experimental data through a theoretical model to verify that any charges developed on the NF membrane’s surface are negligible, validating it as a non-selective membrane.
One challenge in making such a system work with an NF membrane was that these membranes are normally used in filtration for which water flows through them, rather than beside them as the team intended. The team used electrodes patterned with interdigitated flow fields, which were introduced in previous work, to minimize the pumping pressure needed to flow water through the electrodes without causing seepage through the NF membrane.
“It is exciting to see how innovation in fluid mechanics has enabled innovation toward electrochemical functionality,” Smith said. “I didn’t anticipate that ahead of time.”
“We’ve shown that the use of a non-selective membrane is more appropriate for brackish water desalination than for seawater desalination,” Do said. “The next step is to scale up our system.”
Replacing the AEM with an NF membrane also led to increased water recovery, which describes the ratio of freshwater volume produced to total volume of feed water used. Typically, if a system starts with equal volumes of brine and diluate, the desalination process will result in decreased water recovery—i.e., decreased diluate and increased brine due to osmosis happening at the membrane. However, Vu and Smith observed increased diluate volume and decreased brine volume after the desalination process—an unprecedented and unexpected increase in water recovery.
Furthermore, their theoretical analysis has shown that the salt concentration at the NF membrane/brine interface is significantly lower than that at the AEM/brine interface, making the NF membrane less susceptible to fouling—a condition in which salt begins to crystallize on the membrane’s surface, clogging its pores.
This exciting achievement comes full-circle for Smith’s research team, who predicted in a previous publication that IEM elimination would be achievable.
This research was supported by the Expeditionary Energy Program of the US Office of Naval Research (Award no. N00014-22-1-2577).