View source for Item:OSW0f9365f91b664f4c9331f91572321b8f

== '''Agenda''' ==
15:00 -15:05 Intro BATTERY 2030+ Erik Berg

15:05- 16:00 Presentation “Safe and Low-Cost Aqueous Energy Storage Technologies and Their Applications” Yi-Chun Lu

16:00- 16:30 Q&A and Summery Erik Berg

== Recording ==
{{#ev:youtube|QKynDwy_I5Q|||Safe and Low-Cost Aqueous Energy Storage Technologies and Their Applications - Yi-Chun Lu - Battery 2030+ Excellence Seminar}}

== Transcript ==
My name is Yi-Chun Lu, and I'm thankful for this opportunity to share our work. 

Today, I will be mainly talking about aqueous batteries. I thought it would be good to start by introducing an overview of my research activities. As Erik mentioned, we have projects in metal-air and metal-sulfur batteries, where we focus on the fundamental understanding of electrode and electrolyte interfaces, as well as aqueous batteries and redox flow batteries. Today, I will focus mainly on the aqueous batteries and redox flow batteries. 

We’ve built several in-situ in-house characterization systems to understand battery chemistry while operating the batteries. We want to know what kinds of side reactions or degradation mechanisms are ongoing. This is something very close to our hearts. In addition to fundamental science, we also dedicate a lot of effort to technology transfer. We founded Luquos Energy, which focuses on commercializing large-scale energy storage for renewable energy integration.

To begin, I’d like to start with the motivation for the first part of my talk on aqueous batteries. We all know that non-aqueous batteries have high energy density, but they are flammable. This can be risky, especially in applications like electric vehicles (EVs) or large-scale energy storage, where the larger the scale, the greater the risk. Aqueous batteries, on the other hand, are safer but suffer from a narrow voltage window due to the stability limits of water. The voltage window for aqueous batteries is restricted by the oxygen evolution and hydrogen evolution reactions.

There have been efforts to expand the voltage window of aqueous batteries using highly concentrated electrolytes, such as water-in-salt and hydrate-melt systems. These approaches have improved stability but involve high costs and potential toxicity issues due to the high salt concentration. Our goal was to find an approach to stabilize water without relying on highly concentrated salts. Inspired by the phenomenon of molecular crowding in living cells, we wondered if we could apply this concept to aqueous electrolytes to reduce water activity using a crowding agent that interacts strongly with water.

We developed a series of water-miscible polymers that are low-cost and eco-friendly. Our idea was to increase the interaction between the water and the crowding agent, thereby discouraging water-water interactions. This would weaken the hydrogen bond network between water molecules, which would strengthen the O-H covalent bonds and reduce water splitting. To test this, we used polyethylene glycol (PEG), a common crowding agent in biology, which can mix with water in any ratio. We experimented with various PEG and water concentrations to study how electrolyte properties change as PEG content increases. Since these materials are non-toxic and low-cost, they offer significant advantages over lithium salts.

We first examined how the voltage window changes with increasing PEG concentration. As PEG content increases, the hydrogen evolution potential is significantly delayed, suggesting that water splitting is suppressed. We confirmed this using NMR, which showed that higher PEG content weakens the hydrogen bond network, as the strong water-water hydrogen bonds are replaced by weaker water-PEG interactions. This is also supported by a shift in the O-H covalent bond strength, which further strengthens our hypothesis.

Using an electrolyte with 94% PEG and 6% water, we tested lithium manganese oxide (LMO) as the cathode and lithium titanate (LTO) as the anode in a full cell. We also used online electrochemical mass spectrometry to check for side reactions. There was no evidence of hydrogen or oxygen evolution during cycling, confirming that the molecular crowding electrolyte suppresses water splitting. The cell operated for over 300 cycles with more than 80% capacity retention. We compared this low lithium concentration electrolyte with other highly concentrated electrolytes and found that it provided one of the widest potential windows for aqueous electrolytes.

However, one drawback of our first attempt was the high viscosity of the PEG-water electrolyte, which resulted in low ionic conductivity (0.8 milliSiemens per cm). This is lower than water-in-salt systems and other highly concentrated electrolytes, which typically have ionic conductivities below 1 milliSiemens per cm. To address this, we explored other water-miscible polymers with lower viscosities, such as PEGDME, which does not have strong intermolecular hydrogen bond interactions. This reduced the viscosity and improved the ionic conductivity, as well as reducing voltage hysteresis during cycling.

We observed no hydrogen or oxygen evolution with the PEGDME molecular crowding electrolyte, indicating that this concept holds promise for aqueous batteries. In fact, we’ve seen many papers applying molecular crowding electrolytes to various battery systems, such as zinc batteries and supercapacitors.

Next, I’ll shift gears to flow batteries, which are particularly suited for large-scale, long-duration energy storage. Flow batteries allow you to scale the energy content by increasing the size of the external tanks that store the dissolved active materials, without changing the power stack. This decoupling of power and energy makes flow batteries highly scalable for long-duration storage. Additionally, aqueous electrolytes in flow batteries are non-flammable, and self-discharge can be easily controlled by disabling the pump when the battery is idle.

The main challenge with flow batteries, however, is the high cost of the electrolyte. Vanadium is the most mature flow battery chemistry, but it is expensive and limited by low energy density (about 25-30 watt-hours per liter). To address this, we explored sulfur, specifically polysulfide in aqueous systems, which is much cheaper and more abundant than vanadium. In 2016, we developed a polysulfide-iodide flow battery that demonstrated an energy density of around 50 watt-hours per liter. However, the main challenge was the limited cycle life due to crossover of polysulfide and water through the Nafion membrane.

To overcome this, we modified the Nafion membrane by adding a hydrophobic polymer-bound carbon coating. The hydrophobic polymer (PVDF) reduces water migration, while the carbon absorbs polysulfide and polyiodide, preventing their crossover. This simple but effective membrane modification allowed us to significantly improve the cycle life of the polysulfide flow battery. With the modified membrane, we cycled the battery for over three months without degradation, achieving more than 500 cycles and maintaining over 99% capacity.

We also used small-angle X-ray scattering to show that the water channels in the modified membrane were reduced from 3.7 to 2.7 nanometers, which explains the reduced water uptake. Additionally, in-situ FTIR showed that water migration through the membrane was significantly slower compared to the unmodified Nafion membrane.

Another challenge with polysulfide flow batteries is the sluggish kinetics of the polysulfide redox reactions. Breaking the sulfur-sulfur bond is an energy-intensive process, and traditional solid catalysts have limited active sites. Instead, we explored the use of a soluble molecular catalyst, riboflavin sodium phosphate (FMN-Na). This molecule can be reduced at much lower overpotential and can then chemically reduce polysulfide, significantly improving the kinetics of the system. We demonstrated that this molecular catalyst improved both the rate capability and the cyclability of the battery, allowing for higher capacity utilization without reaching the cutoff voltage.

We confirmed the effectiveness of the molecular catalyst by observing the charge transfer between FMN-Na and polysulfide using UV-VIS spectroscopy. The catalyst effect persisted over multiple cycles, and we observed significant improvements in overpotential and energy efficiency. We also scaled up the system and tested it at 100 mA per cm² in a 100 cm² cell, demonstrating stable operation at scale.

Finally, I want to discuss flow batteries for low-temperature applications. Cold weather can limit the performance of flow batteries, so we explored polyoxometalate (POM) as a material for low-temperature operation. POM has high kinetics and can store up to six electrons, making it a good candidate for low-temperature flow batteries. We coupled POM with a vanadium electrolyte and demonstrated stable operation at temperatures as low as -20°C, with high current densities and long cycle life.

In conclusion, we’ve made significant progress in both aqueous and flow battery technologies. For aqueous batteries, molecular crowding electrolytes offer a promising way to stabilize water and expand the voltage window. For flow batteries, we’ve improved cycle life and kinetics by modifying membranes and using molecular catalysts, and we’re exploring new materials for low-temperature applications.

I’d like to thank my funding agencies, students, and postdocs for their contributions. I’m happy to take any questions.