Aqueous Organic Redox Flow Batteries for Grid Energy Storage

The Need for Grid Energy Storage

Fig. 1: Schematic illustration of redox flow battery design; general mechanism which applies to all potential battery chemistries. [1] (Image Source: S. Yribarren)

In order to reach CO₂ reduction goals, we must innovate to "decarbonize" electricity grids. Decarbonization requires that the electrons flowing through power lines are generated by carbon-free (e.g., wind, solar) as opposed to traditional carbon-based (e.g., coal, natural gas) sources. However, a significant challenge to achieving decarbonization is a lack of energy storage. Renewables such as wind and solar are intermittent, producing energy when weather conditions are favorable (sunny days, windy nights) rather than when demand is highest. Widespread energy storage would allow grid operators to store any excess energy generated when supply outpaces demand for example, midday when the sun is shining brightest, or midnight when wind is most powerful. Then, at times when the supply of carbon-free energy is unable to keep up with demand, that stored energy can be sent to the grid, instead of relying on burning carbon-based sources to fill the supply-demand gap.

Flow Battery: General Operating Mechanism

Redox flow batteries (RFBs) are a form of long-duration energy storage that utilize reduction- oxidation (redox) chemistry to reversibly convert electrical to chemical potential. As the schematic in Fig. 1 illustrates, flow batteries have two tanks containing a positive electrolyte and a negative electrolyte. Dissolved in the electrolytes is a redox-active molecule. The fluid from these tanks is pumped into a battery cell, separated by a proton (H+) exchange membrane. During charging, excess energy flowing from the grid charges the battery, pulling electrons from the positive solution (oxidation) and pushing them into the negative solution (reduction). Protons cross the proton exchange membrane (PEM) to keep the system thermodynamically stable. During discharge, when the battery turns on, the electron flow and redox chemistry reverses, and it generates an electric current, which can then be sent to the grid for use.

One benefit of these, for grid storage in particular, is that the system architecture allows independent scaling of energy and power. This is compared to the lithium ion (Li-ion) battery where power-producing and energy-storage components are physically coupled, meaning increasing in the size of the battery increases both power output and energy storage. For RFBs, by contrast, energy can be scaled by increasing the volume of electrolyte storage tanks, and power can be scaled by adding more battery stacks to flow electrolyte by. [1] This decoupled design is ideal for grid storage, which requires fine control over power output (often over long periods), and also needs to accommodate large amounts of stored energy. There are many possible designs for RFBs, but vanadium-based chemistries currently have the highest Technological Readiness Level (TRL). Table 1 uses the Pacific Northwest National Laboratory Energy Storage Cost and Performance Database to compare Li-ion with VRFB design and highlight some upsides and downsides of each battery type for grid storage, using 2021 data. [2]

The comparison shows a number of benefits of flow compared to Li-ion batteries, for grid energy storage in particular. Redox flow batteries have a comparable overall calendar life to Li-on, but virtually unlimited cycle-life, so can be more active throughout its commission period. They need less rest before charge/discharge which increases flexibility as needed by grid operators. Additionally, they are also safer: cross-mixing of positive and negative electrolytes will not result in explosions and they do not pose a fire hazard, plus they can operate at lower (and for some designs even room) temperatures. [4] Lastly, while Li-ion is more energy- and power-dense, the stationary nature of grid storage means this matters less than other energy storage applications, and thus does not require lithium or other energy-dense elements (though they might be beneficial in limiting land-use footprint). This means many different chemical combinations can operate a functional flow battery, including abundant (and often cheaper) organic and inorganic molecules.

However, despite the theoretical benefits, the final Levelized Cost of Storage (LCOS) line demonstrates that cost remains a major issue in deployment, and is why the technology is not seen at a widespread commercial scale today. The LCOS is not competitive with Li-ion batteries, with Lithium iron phosphate (LFP) technology costing about 77.8% of vanadium-ion battery technology as seen in Table 1. This is a result of higher initial capital cost, higher operating costs, and higher decommissioning costs. While the invention is technologically mature, it is clear that VRFB struggles to be cost-competitive with Li-ion. This has prevented it from scaling, and thus it remains largely irrelevant as a technology, practically-speaking. Thus, new redox-flow battery chemistries are needed to overcome the downsides of vanadium-based chemistry.

What is clear from this historical case study is that high technological readiness level (TRL) does not result in economic viability, and thus scalability and impact. There were fundamental problems from the beginning of VRFB development that led to its failure today, which I outline below. In the rest of this report, I compare VRFBs with a novel aqueous-organic redox-flow battery chemistry to consider systems-level factors that could lead to greater economic viability than VRFBs if a high TRL is able to be achieved.

Traditional Design: Vanadium Ions in Acidic Solution

In a previous post on this website from 2011, a student analyzed the Vanadium Redox-Flow Battery, which at the time appeared the most promising chemistry for flow batteries. [5] The Department of Energy has funded many projects for VRFBs over the years, which has improved the chemistry but ultimately failed its quest of widespread scale-up due to economic viability concerns. This chemistry used vanadium ions in various oxidation states, and results in ~1.25 V:

For this battery design, the components and associated concerns are:

Table 1 illustrated the overall cost concerns with VRFBs. Table 2 breaks down each main component of the VRF batteries, and the specific physical concerns that contribute to its economic infeasibility. The vanadium minerals used are expensive, due to the mining and processing needed, and the price is volatile due to its association with steel production. [6,7] The graphite felt electrodes are chosen for chemical stability but, given their operation in low pH (~2-4) conditions, degrade overtime and add to lifetime cost for the cells. [8] The use of sulfuric acid (H₂SO₄) to dissolve V-ions necessitates special corrosive-resistant equipment which is often expensive, and needs periodic replacement. [8] And lastly, the Dupont-made proton exchange membrane Nafion is industry-standard with V-ions due to their small size, which makes selectivity between V-ions and protons more difficult without highly specialized membranes. [9] However, the membrane alone can contribute more than 40% of the overall battery cost. [5,9]

Solving these physical problems through developing a new flow battery design (not reliant on vanadium ions) could remedy associated cost concerns.

Newer Design: Redox-Active Organics in Aqueous Solution

To retain the benefits of flow batteries compared to Li-ion, but avoid the design flaws of the vanadium-based redox chemistry, one of many newly developed chemistries uses redox-active organics in place of vanadium ions, and dissolves them in water instead of concentrated sulfuric acid. [7] This type of battery is called an Aqueous Organic Flow Battery. There is a theoretically infinite amount of organic chemical species you can utilize in this design (e.g. quinones, viologens, nitroxide radicals, aza-aromatics, iron coordination complexes have all been used) and companies in France, Italy, Germany, and the U.S. are all utilizing different battery chemistries to attempt commercial scale-up. Because there is no standardized chemistry for AORFBs, in order to highlight differences between AORFBs and VRFBs, I will focus on the specific design provided by US-based company Quino Energy. I am choosing this one in particular due to its current [Oct. 2024] rating at Technology Readiness Level 7 (TRL7) and Manufacturing Readiness Level 6 (MRL6), which to my knowledge is the highest of all current AORFB companies. [10]

As seen with the failure of vanadium flow batteries, having a high TRL does not mean a technology will be economically viable. Thus, it is important to consider broader systems factors that might prevent scalability before resources and effort are spent on technological development. This is an attempt to do that through comparison to VRFBs. The battery chemistry uses ferrocyanide for the positive electrolyte, and a quinone called DHAQ (2,6-dihydroxy-anthraquinone) for the negative electrolyte. [11] Both are abundant plant-based organic compounds, a food additive and dye respectively. The overall reaction in the cell results in ~1.22 V: [11]

This leads to complex network of reactions seen below. Overall the conversion between the compounds above (which releases or gains two electrons) is around E₀ = -0.77 V. [11] (See also Fig. 3)

This 1.22 V voltage output is very similar to the 1.25 V from the V-ion RFB. For this battery design, each basic component and their associated concerns are:

This analysis of this one in-development AORFB design illuminates the potential systems-level benefits of aqueous-organic over acid-vanadium flow battery design. Most importantly, it shows how AORFBs might avoid the economic challenges of VRFBs through the alternate design. By replacing vanadium ions with abundant and cheap organic molecules as the redox-active compound, AORFBs could avoid expensive prices and market volatility. Dissolving in water instead of sulfuric acid takes away the cost of the acid, the need for expensive equipment, as well as corrosivity concerns for the electrodes and membrane. Also, since the organic compounds are bulkier than vanadium ions, there is potential to switch from the expensive Nafion ion-exchange membranes to a size-selective filtration membrane instead, which could save a lot on cost. [12] Further, redox-active organics are compatible across a wide pH range, though slightly alkaline conditions are preferable. [10]

Concerns with this (and most other) AORFB designs are almost entirely to do with the chemical stability of the organic species, which are known to degrade. [13] For ferrocyanide, it is stable across time, but typically degrades in high-alkaline environments, though some recent studies shed doubt on this claim. [10,14,15] Therefore, pH swings need to be closely monitored and more efficient pumping might be required, which could lead to higher operating costs. [10,17] For quinones, though they are common in lab-scale AORFB setups, generally they degrade overtime which is a problem for commercial scale-up. [16] If the molecule degrades in the storage tanks, the stored chemical energy is no longer accessible. Quino Energy claims to have invented a chemical system (a dimerization network distilled above) where chemical stability of quinones can be maintained over long periods without degradation. [11] This supposed breakthrough appears to be why they are at a higher TRL and MRL than other AORF battery companies. However, the validity of this claim will become evident with time as more of their batteries come online and undergo real-world testing. Without the claimed chemical stability, the technical and economic feasibility of Quino's batteries are greatly diminished and highly unlikely to scale.

Conclusion

The physical and engineering principles underlying vanadium-acid and aqueous-organic flow batteries are the same, as shown in Fig. 1. However, they utilize different underlying chemistries to achieve different results for cost, performance, and safety. As discussed and shown in Table 1, flow batteries are poised to be a better energy storage method for grid applications overall compared to Li-ion, if it can be made economically viable. Yet Table 2 shows that vanadium RFBs have various concerns related to capital and operating costs, as well as safety and price volatility. Table 3 illustrates that aqueous organic flow batteries present an alternative, replacing corrosive acid with water, expensive redox-active ions with more abundant ones, and more. Without conducting an in-depth techno-economic assessment of a specific design, this surface analysis indicates that the overall principles behind organic flow batteries might be able to lower the levelized cost of storage for flow batteries to be lower than Li-ion.

The data to prove this point is not yet available at a comparable scale, and is not yet even under consideration in the PNNL database (used for comparison in Table 1) since AORFBs are still a nascent, in- development energy storage technology. What is evident is that there does not appear to be any major red flags that would hinder their economic feasibility down the line, though chemical stability of organic species is a yellow flag to watch out for. This is in contrast to the vanadium case study, where the problems outlined in Table 2 (such as scarcity/volatility of vanadium) should have been evident from the outset, but a long-term view of economic viability for scalability was not taken.

Right now, it appears clear that the initial capital costs of AORFBs are significantly less than VRFBs due to lower equipment cost. However, one modeling framework suggests that the low capital costs initially may end up being more than offset by the high operating costs over the lifetime of the battery, perhaps due to higher pumping costs to control pH, as previously mentioned.[10,17] Figuring out why the operating costs might be higher can help to focus efforts on designs with the most potential, accelerating the development of grid-scale batteries towards decarbonization goals.

© Sarah Yribarren. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.

References

[1] H. Zhang, X. Li, and J. Zhang, eds. Redox Flow Batteries: Fundamentals and Applications (CRC Press, 2017).

[2] V. Viswanathan et al., "2022 Grid Energy Storage Technology Cost and Performance Assessment," Pacific Northwest National Laboratory, PNNL-333283, August 2022.

[3] X. Luo et al., "Overview of Current Development in Electrical Energy Storage Technologies and the Application Potential in Power System Operation," Appl. Energy 137, 511 (2015)

[4] M. Uhrig et al., "Lithium-Based vs. Vanadium Redox Flow Batteries - A Comparison for Home Storage Systems," Energy Procedia 99, 35 (2016).

[5] X. Xie, "Vanadium Redox-Flow Battery," Physics 240, Stanford University, Fall 2011.

[6] D. Emmel et al., "Benchmarking Organic Active Materials For Aqueous Redox Flow Batteries in Terms of Lifetime and Cost," Nat. Commun. 14, 6672 (2023).

[7] M. L. Perry, K. E. Rodby, and F. R. Brushett, "Untapped Potential: The Need and Opportunity for High-Voltage Aqueous Redox Flow Batteries," ACE Energy Lett. 7, 659 (2022).

[8] H. Liu et al., "Corrosion Behavior of a Positive Graphite Electrode in Vanadium Redox Flow Battery," Electrochim. Acta 56 8783 (2011).

[9] A. Trovò et al., "Redox Flow Batteries: A Glance at Safety and Regulation Issues," Electronics 12, 1844 (2023).

[10] F. R. Brushett, M. J. Aziz and K. Rodby, "On Lifetime and Cost of Redox-Active Organics for Aqueous Flow Batteries," ACS Energy Lett. 5, 879 (2020).

[11] Y. Jing et al., "In Situ Electrochemical Recomposition of Decomposed Redox-Active Species in Aqueous Organic Flow Batteries," Nat. Chem. 14, 1103 (2022).

[12] Y. Jing, R. G. Gordon and M. J. Aziz, "Aqueous Organic Flow Batteries," in Redox Flow Batteries. From Fundamentals to Applications, Vol. 2, ed. by C. Roth, J. Noack, and M. Skyllas-Kazacos (Wiley-VCH, 2023).

[13] F. Hasan, V. Mahanta, and A. A. A. Abdelazeez, "Quinones for Aqueous Organic Redox Flow Battery: A Prospective on Redox Potential, Solubility, and Stability," Adv. Mater. Interfaces 10, 1300268 (2023).

[14] C. A. P. Arellano and S. S. Martinez, "Effects of pH on the Degradation of Aqueous Ferricyanide by Photolysis and Photocatalysis Under Solar Radiation," Sol. Energy Mater. Sol. Cells 94, 327 (2010).

[15] E. M. Fell et al., "Long-Term Stability of Ferri-/Ferrocyanide as an Electroactive Component for Redox Flow Battery Applications: On the Origin of Apparent Capacity Fade," J. Electrochem. Soc. 170, 00525 (2023).

[16] D. P. Tabor et al., "Mapping the Frontiers of Quinone Stability in Aqueous Media: Implications for Organic Aqueous Redox Flow Batteries," J. Mater. Chem. A 7, 12833 (2019).

[17] K. E. Rodby, M. L. Perry, and F. R. Brushett, "Assessing Capacity Loss Remediation Methods For Asymmetric Redox Flow Battery Chemistries Using Levelized Cost of Storage," Journal of Power Sources 506, 230085 (2021).