HIGHLIGHTS
• A derivative of a byproduct from triphenylphosphine oxide, known as CPO, has been found to be a suitable anolyte for use in nonaqueous redox flow batteries.
• The stability of CPO under operating conditions in the redox flow battery is due to a conjugated system that is developed in its cycle structure.
• Stable galvanostatic cycling without showing any fade was accomplished in performance testing when CPO was used in a specific blend ratio of the two solvents dimethylformamide and acetonitrile.

The growing demand for electrification, particularly with regard to data centers needed to meet higher digital needs, is leading to finding new approaches for generating energy renewably. With the vulnerability of utilizing wind and solar to produce electricity only under certain environmental conditions, steps are underway to figure out how to store energy for times when the sun is not available and the wind is not blowing.

One option that is under development is utilization of redox flow batteries. Aqueous redox flow batteries were initially studied but have challenges because they can only provide an operating voltage of 1.23 V and have limited energy densities. 

In a previous TLT article,¹ research on a new type of non-aqueous redox flow battery was discussed. The battery was prepared with an anolyte (material situated near the battery anode and reduced during battery charging) and a catholyte (material situated near the battery cathode and oxidized during battery charging) that were not compatible with each other. This eliminated concerns about crossover that could impact performance. Evaluation of the concept in a battery led to a retention rate of 81% over 100 cycles with an average cell voltage of 3.45 V and a current density of 1.5 milliamps per centimeter. The testing took 45 days to complete, and the researchers predict that such a battery can operate for at least one year without any difficulties.

To make nonaqueous redox flow batteries a more sustainable concept consistent with their role in storing energy, materials that can be used in the anolyte and/or the catholyte need to be identified that are renewable or are sourced from a waste byproduct that can be upcycled in some fashion. One such material is triphenylphosphine oxide (TPPO). 

Dr. Christian Malapit, assistant professor of chemistry at Northwestern University in Evanston, Ill., says, “Past work in identifying electrolytes for nonaqueous redox flow batteries has focused on carbonyl and pyridinyl derivatives. But these options are not sustainable and usually need to be prepared in multi-step synthesis pathways, which increases their cost. Our approach was to identify a material that was readily available, low cost and could be converted easily into a useful electrolyte.”

TPPO was found to be a viable candidate because multi-metric ton quantities of this material are generated in industrial and laboratory processes. Malapit says, “TPPO is attractive as a potential anolyte because it can be reduced to a high negative potential. The problem is TPPO in a reduced state is unstable. One additional factor from a sustainability standpoint is recycling TPPO after it is used is not economically viable. TPPO can be converted to triphenyl phosphine (TPP), but this is a difficult process to carry out and is energy intensive.”

Malapit and his colleagues decided the next best option was to molecularly engineer a derivative of TPPO that exhibits a similar high negative potential but is also more stable under the operating condition needed in a nonaqueous redox flow battery. Such a derivative has now been synthesized and evaluated.

CPO
The researchers found that reacting TPPO with 5-phenylbenzo[b] phosphindole 5-oxide led to the synthesis of derivative known as CPO. Malapit says, “We determined that CPO displayed a high negative potential that is reversible in two of the most common solvents used in nonaqueous redox flow batteries, acetonitrile and dimethylformamide. The synthesis of CPO from TPPO can be accomplished in a one-pot reaction and the pure derivative can be isolated after a simple recrystallization (see the top process in Figure 3).”


Figure 3. A potential anolyte that can be used in a nonaqueous redox flow battery is known as CPO and can be prepared from the byproduct triphenyl phosphine oxide (top process). The charging step for CPO involves conversion to a radical anion during a reduction reaction (bottom process). Figure courtesy of Northwestern University.

The next step in the process for the researchers was to isolate and evaluate the radical anion of CPO produced during reduction (charging—see the bottom process in Figure 3). X-ray crystallographic analysis indicated that the CPO structure remained intact with no bond modification after reduction. Electron paramagnetic spectroscopy demonstrated that the radical character seems to be mainly delocalized on the backbone.

Malapit says, “The reason for the stability of CPO is due to a conjugated system that is developed in its cycle structure. The presence of two aromatic rings attached to each other in a planar orientation stabilizes the molecule.”

A static H-cell was used to evaluate the performance of CPO in charge and discharge experiments (see Figure 4). During initial testing, the researchers found that CPO survived for over 100 cycles in acetonitrile and dimethyl formamide but significant capacity fade was encountered. With crossover ruled out after analysis of the anolyte and catholyte solutions, the researchers analyzed the post-anolyte solution using ³¹P nuclear magnetic resonance spectroscopy.


Figure 4. The performance of CPO in charge and discharge experiments was evaluated using a static H cell. Figure courtesy of Northwestern University.

Malapit says, “We found that CPO was reduced to a cyclic phosphine in acetonitrile and decomposed to an unknown phosphorus-containing compound in dimethylformamide. The two solvents displayed complementary properties as the speed of the process (kinetics) with acetonitrile is better and the reduction window with dimethylformamide is better.”

The researchers figured that evaluating mixtures of the two solvents may be the best approach for finding the right trade-off to maximize performance and minimize potential performance issues. After a series of experiments, a blend of 60% dimethylformamide and 40% acetonitrile produced stable galvanostatic cycling without showing any fade after 350 cycles. 

Research is continuing to increase the stability of this new anolyte candidate. Malapit says, “We are evaluating polymeric versions of CPO where there may be increased electron density combined with better stabilization. Slurry polymer systems will also be evaluated. We will also assess the possibility of potential crossover occurring.”

Additional information can be found in a recent paper² or by contacting Malapit at christian.malapit@northwestern.edu.
REFERENCES
1. Canter, N. (2024), “Non-aqueous redox flow battery,” TLT, 80 (1), pp. 18-19. Available at www.stle.org/files/TLTArchives/2024/01_January/Tech_Beat_II.aspx.
2. Mahoney, E., Boudjelel, M., Shavel, H., Krzyaniak, M., Wasielewski, M. and Malapit, C. (2025), “Triphenylphosphine oxide-derived anolyte for application in nonaqueous redox flow battery,” J. Am. Chem. Soc., 147 (2), pp. 1381-1386.