Tesla provides customers with upgraded charging software and adapter to reduce potential risk of garage fires
Petrobras launches new 50 ppm ultra-low-sulfur gasoline throughout Brazil

Harvard team demonstrates new metal-free organic–inorganic aqueous flow battery; potential breakthrough for low-cost grid-scale storage

Cell schematic. Discharge mode is shown; the arrows are reversed for electrolytic/charge mode. AQDSH2 refers to the reduced form of AQDS. Huskinson et al. Click to enlarge.

Researchers at Harvard have demonstrated a metal-free organic–inorganic aqueous flow battery—a quinone–bromide flow battery (QBFB)—as an example of a class of energy storage materials that exploits the favorable chemical and electrochemical properties of a family of molecules known as quinones. Quinones are naturally abundant, inexpensive, small organic molecules, and similar to molecules that store energy in plants and animals. The new flow battery developed by the Harvard team already performs as well as vanadium flow batteries, with chemicals that are significantly less expensive and with no precious-metal electrocatalyst.

In a paper in Nature, they suggest that the use of such redox-active organic molecules instead of redox-active metals represents a new and promising direction for realizing massive electrical energy storage at greatly reduced cost. The technology could fundamentally transform the way electricity is stored on the grid, making power from renewable energy sources such as wind and sun far more economical and reliable.

As the fraction of electricity generation from intermittent renewable sources—such as solar or wind—grows, the ability to store large amounts of electrical energy is of increasing importance. Solid-electrode batteries maintain discharge at peak power for far too short a time to fully regulate wind or solar power output. In contrast, flow batteries can independently scale the power (electrode area) and energy (arbitrarily large storage volume) components of the system by maintaining all of the electro-active species in fluid form. Wide-scale utilization of flow batteries is, however, limited by the abundance and cost of these materials, particularly those using redox-active metals and precious-metal electrocatalysts.

—Huskinson et al.

Background. Flow batteries store energy in chemical fluids contained in external tanks, as with fuel cells, instead of within the battery container itself. The two main components—the electrochemical conversion hardware through which the fluids are flowed (which sets the peak power capacity) and the chemical storage tanks (which set the energy capacity)—may be independently sized. Thus the amount of energy that can be stored is limited only by the size of the tanks. The design permits larger amounts of energy to be stored at lower cost than with traditional batteries.

By contrast, in solid-electrode batteries such as those commonly found in cars and mobile devices, the power conversion hardware and energy capacity are packaged together in one unit and cannot be decoupled. Consequently they maintain peak discharge power for less than an hour before they are drained, and are therefore ill-suited to store intermittent renewables.

Michael J. Aziz, the Gene and Tracy Sykes Professor of Materials and Energy Technologies at the Harvard School of Engineering and Applied Sciences (SEAS), who led the design, building and testing of the battery, notes that his lab’s studies indicate that one to two days’ worth of storage is required for making solar and wind dispatchable through the electrical grid.

To store 50 hours of energy from a 1-megawatt power capacity wind turbine (50 megawatt-hours), for example, a possible solution would be to buy traditional batteries with 50 megawatt-hours of energy storage, but they would come with 50 megawatts of power capacity. Paying for 50 megawatts of power capacity when only 1 megawatt is necessary makes little economic sense.

For this reason, a growing number of engineers have focused their attention on flow-battery technology. But until now, flow batteries have relied on chemicals that are expensive or hard to maintain, driving up the cost of storing energy.

The active components of electrolytes in most flow batteries have been metals. Vanadium is used in the most commercially advanced flow-battery technology now in development, but it sets a rather high floor on the cost per kilowatt-hour at any scale. Other flow batteries contain precious metal electrocatalysts, such as the platinum used in fuel cells.

The Harvard metal-free organic–inorganic aqueous flow battery. The example reported in the Nature paper is a metal-free flow battery based on the redox chemistry of 9,10-anthraquinone-2,7-disulphonic acid (AQDS). The organic anthraquinone species can be synthesized from inexpensive commodity chemicals. AQDS undergoes extremely rapid and reversible two-electron two-proton reduction on a glassy carbon electrode in sulfuric acid.

Solutions of AQDS in sulphuric acid (negative side) and Br2 in HBr (positive side) were pumped through the flow cell; the quinone–bromide flow battery (QBFB) was constructed using a Nafion 212 membrane sandwiched between Toray carbon paper electrodes (six stacked on each side) with no catalysts.

The QBFB yields a peak galvanic power density exceeding 0.6 W cm−2 at 1.3 A cm−2. Cycling of this quinone–bromide flow battery showed more than 99% storage capacity retention per cycle.

The organic approach liberates battery redox chemistry from the constraints of the limited number of elemental redox couples of the periodic table. Although quinones have been used previously in batteries using redox-active solids, their incorporation into all-liquid flow batteries offers the following advantages over current flow-battery technologies. First, scalability: AQDS contains only the Earth-abundant atoms carbon, sulphur, hydrogen and oxygen, and can be inexpensively manufactured on large scales. Because some hydroxy-anthraquinones are natural products, there is also the possibility that the electrolyte material can be renewably sourced.

Second, kinetics: quinones undergo extremely rapid two-electron redox on simple, inexpensive carbon electrodes and do not require a costly precious-metal catalyst. Furthermore, this electrode permits higher charging voltages by suppressing the parasitic water-splitting reactions. Third, stability: quinones should exhibit minimal membrane crossover owing to their relatively large size and charge in aqueous solution as a sulphonate anion. Furthermore, although bromine crossover is a known issue in zinc-bromine, vanadium-bromine and hydrogen-bromine cells, AQDS is stable to prolonged heating in concentrated Br2/HBr mixtures, and the QBFB can be cycled in HBr electrolyte solutions. Fourth, solubility: AQDS has an aqueous solubility greater than 1 M at pH 0, and the quinone solution can thus be stored at relatively high energy density—volumetric and gravimetric energy densities exceed 50 W h l-1 and 50 W h kg-1, respectively. Last, tunability: the reduction potential and solubility of AQDS can be further optimized by introduction of functional groups such as –OH.

Use of DHAQDS is expected to lead to an increase in cell potential, performance and energy density. These features lower the capital cost of storage chemicals per kilowatt hour, which sets a floor on the ultimate system cost per kilowatt hour at any scale. The precursor molecule anthracene is abundant in crude petroleum and is already oxidized on large scale to anthraquinone. Sulphonated anthraquinones are used on an industrial scale in wood pulp processing for paper, and they can be readily synthesized from the commodity chemicals anthraquinone and oleum.

…Based on this simple electrolyte preparation that requires no further product separation, we estimate chemical costs of $21 per kilowatt hour for AQDS and $6 per kilowatt hour for bromine… The QBFB offers major cost improvements over vanadium flow batteries with redox-active materials that cost $81 per kilowatt hour. Optimization of engineering and operating parameters such as the flow field geometry, electrode design, membrane separator and temperature—which have not yet even begun—should lead to significant performance improvements in the future, as it has for vanadium flow batteries, which took many years to reach the power densities we report here. The use of redox processes in π-aromatic organic molecules represents a new and promising direction for cost-effective, large-scale energy storage.

—Huskinson et al.

The organic approach also permits tuning of important properties such as the reduction potential and solubility by adding functional groups: for example, the addition of two hydroxy groups to AQDS increases the open circuit potential of the cell by 11%; the Harvard team also describes a pathway for further increases in cell voltage.

Quinones are abundant in crude oil as well as in green plants. The molecule the Harvard team used in its first quinone-based flow battery is almost identical to one found in rhubarb. The quinones are dissolved in water, which prevents them from catching fire.

The whole world of electricity storage has been using metal ions in various charge states, but there is a limited number that you can put into solution and use to store energy, and none of them can economically store massive amounts of renewable energy. With organic molecules, we introduce a vast new set of possibilities. Some of them will be terrible and some will be really good. With these quinones we have the first ones that look really good.

—Roy G. Gordon, the Thomas Dudley Cabot Professor of Chemistry and Professor of Materials Science, who led the work on the synthesis and chemical screening of molecules

To back up a commercial wind turbine, a large storage tank would be needed, possibly located in a below-grade basement, said co-lead author Michael Marshak, a postdoctoral fellow at SEAS and in the Department of Chemistry and Chemical Biology. With a whole field of turbines or a large solar farm, one could imagine a few very large storage tanks. The same technology could also have applications at the consumer level, Marshak said.

Alán Aspuru-Guzik, professor of chemistry and chemical biology, used his pioneering high-throughput molecular screening methods to calculate the properties of more than 10,000 quinone molecules in search of the best candidates for the battery. He noted that the project is very well aligned with the White House Materials Genome Initiative.

This project illustrates what the synergy of high-throughput quantum chemistry and experimental insight can do. In a very quick time period, our team homed in to the right molecule. Computational screening, together with experimentation, can lead to discovery of new materials in many application domains.

—Alán Aspuru-Guzik

Team leader Aziz said the next steps in the project will be to further test and optimize the system that has been demonstrated on the benchtop and bring it toward a commercial scale. He also expects to achieve significant improvements in the underlying chemistry of the battery system.

So far, we’ve seen no sign of degradation after more than 100 cycles, but commercial applications require thousands of cycles. I think the chemistry we have right now might be the best that’s out there for stationary storage and quite possibly cheap enough to make it in the marketplace, but we have ideas that could lead to huge improvements.

—Michael J. Aziz

Commercialization. By the end of the three-year development period, Connecticut-based Sustainable Innovations, LLC, a collaborator on the project, expects to deploy demonstration versions of the organic flow battery contained in a unit the size of a horse trailer. The portable, scaled-up storage system could be hooked up to solar panels on the roof of a commercial building, and electricity from the solar panels could either directly supply the needs of the building or go into storage and come out of storage when needed.

Sustainable Innovations anticipates playing a key role in the product’s commercialization by leveraging its ultra-low-cost electrochemical cell design and system architecture already under development for energy storage applications.

You could theoretically put this on any node on the grid. If the market price fluctuates enough, you could put a storage device there and buy electricity to store it when the price is low and then sell it back when the price is high. In addition, you might be able to avoid the permitting and gas-supply problems of having to build a gas-fired power plant just to meet the occasional needs of a growing peak demand.

—Michael J. Aziz

The technology could also provide very useful backup for off-grid rooftop solar panels—an important advantage considering some 20% of the world’s population does not have access to a power distribution network.

The intermittent renewables storage problem is the biggest barrier to getting most of our power from the sun and the wind. A safe and economical flow battery could play a huge role in our transition off fossil fuels to renewable electricity. I’m excited that we have a good shot at it.

—Michael J. Aziz

William Hogan, Raymond Plank Professor of Global Energy Policy at Harvard Kennedy School and one of the world’s foremost experts on electricity markets, is helping the team explore the economic drivers for the technology.

Trent M. Molter, president and CEO of Sustainable Innovations, LLC, provides expertise on implementing the Harvard team’s technology into commercial electrochemical systems.

The Harvard team received funding from the US Department of Energy’s Advanced Research Projects Agency — Energy (ARPA-E) OPEN 2012 program to develop the grid-scale battery, and plans to work with the agency to catalyze further technological and market breakthroughs over the next several years.

The Harvard team’s results published in Nature demonstrate an early, yet important technical achievement that could be critical in furthering the development of grid-scale batteries. The project team’s result is an excellent example of how a small amount of catalytic funding from ARPA-E can help build the foundation to hopefully turn scientific discoveries into low-cost, early-stage energy technologies.

—ARPA-E Program Director John Lemmon


  • Brian Huskinson, Michael P. Marshak, Changwon Suh, Süleyman Er, Michael R. Gerhardt, Cooper J. Galvin, Xudong Chen, Alán Aspuru-Guzik, Roy G. Gordon & Michael J. Aziz (2014) “A metal-free organic–inorganic aqueous flow battery”Nature 505, 195–198 doi: 10.1038/nature12909



..and we owe it all to rhubarb..


This development was also covered at Science Daily, with this interesting tidbit:

"Imagine a device the size of a home heating oil tank sitting in your basement. It would store a day's worth of sunshine from the solar panels on the roof of your house, potentially providing enough to power your household from late afternoon, through the night, into the next morning...."

That's a mighty large device to store one day's worth of energy.  People often fill heating oil tanks once per season; 500 gallons will run a 100,000 BTU/hr furnace for more than 3 weeks continuously.  Going RE all the way is far more difficult than most people appreciate.


"we estimate chemical costs of $21 per kilowatt hour for AQDS and $6 per kilowatt hour for bromine"


Is it the 5-5-5 battery that everybody wait for. They don't talk using this in a bev, probably because the bev experiment cynetic energy with acceleration, breaking, turning and bosses on the road contrary to a static energy storage configuration.


This could become one of the possible technology for large expandable stationary storage units for grid stabilisation and intermittent power sources.

Storing unused or very low cost e-energy for peak loads and rainy days may become a profitable business sometime in the next decade?

Eventually, it may become a way to have electric locomotives or street cars without continuous overhead cables.


It would work well for solar where you size the panels to provide enough power for a full day in Winter.
However, you would still have to be connected to the grid, in case you got a run of dull days in winter. (Which is very likely).

So adding storage like this to a solar system, might increase its utility from (say) 30% to maybe 90%, but you'll never get it to 100% (without a generator) or the gird.


This isn't even the 1-1-1 battery; something the size of a multi-hundred gallon fuel tank to store a few tens of kW is not something you can put in a vehicle.  But a cheap way to store electricity on daily cycles, very likely.

I have to admit, this would make RE much more practical.  With a big, cheap battery you can absorb overproduction and leave combustion-powered generators cold until needed.  This would eliminate most of the variability-related inefficiencies in the current scheme.


A couple 120 gallon tanks (hot water sized) to store 20 kWh for a 4000 watt solar panel system would charge an EV at night to provide 60 miles of range each day. No more gas stations.


I wouldn't like a spill of Br2 in my house...
I would prefer a H2-fuel cell.
Can store energy for months and use it as backup electricity and to fill my car.
Only waiting for Pt-free catalists...


Lithium bromide has been used for decades in natural gas air conditioners at homes. You would rather have large tanks of compressed hydrogen and oxygen.

Roger Pham

Good point, Alain.
Also, what is the round-trip efficiency of this flow battery?
At 50Wh/kg, it is not much better than lead-acid battery, although at 50Wh/liter, it is definitely worse than lead-acid volumetric-wise, since lead is quite heavy!

H2 can store energy for a lot less than $27/kWh. Geologic large-scale storage of H2 may cost a few pennies/kWh for seasonal-scale storage. H2 tank made from CF costs about ~16-18/kWh, with 1500Wh/kg gravimetric energy density.

Jim McLaughlin

"more than 99% storage capacity retention per cycle".

If that means 99.5%, then after 100 cycles you have lost 40% of the capacity?

It is not clear what degrades, hopefully the fluid, not the rest of the system?


Why not BEV application? 50 W/kg gives us 20kg/kWh or 200 kg/10 kwh or 2000 kg/100 kWh. That is quite acceptable figures since you could place electrolyte tanks whereever you want and whichever shape you want within vehicle utilizing all empty spaces. Even Tesla could consider such option. In case it is problematic recapturing braking energy one could use capacitors or simple hydraulic system. Morevore I believe it's not the end of power density advance.


Jim MacLaughlin,
Article states that no degradation after 100 cycles. But even it is so I believe you can refill electrolyte and recover properties.


Energy efficiency is only ~60%. Coulombic efficiency is only 99.2%, but irreversible side reactions must be limited to 0.002% per cycle to achieve 10,000 cycles. Nice concept, but need proof of durability prior to scale-up.


I'm wondering about the temperature tolerance of these chemicals: If you could put the tanks for this storage system outside the house or bury them in the ground the size of the system wouldn't matter so much. Cost becomes the only real factor.


Dsnydacker, help me out here. I don't really understand what you're saying. You say coulombic efficiency is 'only' 99.2%, but that would seem to be very good, in line with the best values for existing battery types.

But where does this 'energy efficiency' of 60% come from? I thought a battery could have 2 types of losses: coulombic and ohmic. Coulombic is constant, but ohmic depends on the charge discharge rate. And so the statement '60% efficiency' begs for the question: at what rate?

Am I missing something?


You could probably use a couple of used LEAF or Volt packs at home for EV charging. It depends on the price, cycles left at what capacity and recycle value.


About this discrepency between coulombic efficiency and electrical efficiency. You need energy to charge this battery, or "run the ball uphill" and it is likely 40% is wasted in the form of heat. Getting the potential energy out in the form of current is probably not so bad, as the acidity of HBr probably drives the H+ ions across the membrane with great efficiency.

I think the next step is to incorporate LeChatelier's principle or whatnot, to produce the energy within the electolytes via chemical reaction, no generator needed. Simply drop a catalyst or non-catalyst in the hydroquinone side, release the hydrogen, and watch the solute precipitate out. For all we know such a chemical may be nothing more than carbon dioxide or monoxide, which may do nothing more than change the shape of the quinone-like substrate, thereby controlling the absorption or release of the hydrogen.


Energy efficiency of Li-ion cells can exceed 90%.  Maybe this quinone-based flow battery is less efficient, but I don't see any firm data to that effect.


Does it matter either way? This battery is for stationary applications, efficiency and size are less important than cost.

The comments to this entry are closed.