Eni reports first oil from giant Kashagan field in North Caspian Sea
Going Green selects AllCell as Li-ion battery supplier for electric scooters

GWU researchers introduce new class of molten air batteries; significantly greater energy capacity than Li-air

Generalized form of the molten air battery. Licht et al. Click to enlarge.

Researchers at George Washington University led by Dr. Stuart Licht have introduced the principles of a new class rechargeable molten air batteries that offer amongst the highest intrinsic electric energy storage capabilities.

In a paper just accepted and published online by the RSC journal Energy & Environmental Science, Licht and his colleagues show three examples of the new battery’s electron transfer chemistry. These are the iron, carbon and VB2 molten air batteries with respective intrinsic volumetric energy capacities of 10,000 (for Fe to Fe(III)); 19,000 (C to CO32-) and 27,000 Wh liter-1 (VB2 to B2O3 + V2O5), compared to 6,200 Wh liter-1 for the lithium-air battery.

In 2008 a zirconia stabilized VB2 air battery was presented. [Earlier post.] This 11e- (eleven electron) per molecule, room temperature, aqueous electrolyte battery has the highest volumetric energy capacity for a battery, with an intrinsic capacity greater than that of gasoline and an order of magnitude higher than that of conventional lithium ion batteries. The challenge has been to recharge the battery; that is to electrochemically reinsert 11e- into the battery discharge products. Here, this challenge is resolved through the introduction of a new class of molten air batteries.

...Unlike prior rechargeable molten batteries, the [molten air] battery is not burdened by the weight of the active chargeable cathode material. The rechargeable molten air electrode instead uses oxygen directly from the air to yield high battery capacity. This electrode will be shown to be compatible with several high capacity multiple electron redox couples. Three demonstrated new batteries chemistries are the metal (iron), carbon and VB2 molten air batteries with intrinsic volumetric energy capacities greater than that of the well known lithium air battery due to the latter’s single electron transfer and low density limits.

—Licht et al.

As an example of the process, during the charging of the iron molten battery, iron oxide is converted to iron metal via a three-electron reduction, and oxygen is released to the air. During discharge, iron metal is converted back to iron oxide. Li2CO3, which melts at 723°C, and lower melting carbonate eutectics are effective electrolytes, the researchers found. Simple steel foil cathodes and nickel foil anodes are effective for either iron oxide or carbon dioxide splitting.

Iron has been widely explored for battery storage due to its availability as a resource and its capability for multiple electron charge transfer. Retention of the intrinsic anodic storage capacity of these batteries has been an ongoing challenge...In 2010 we introduced the molten carbonate electrolytic conversion of iron oxide to iron as a CO2-free synthesis alternative to the conventional greenhouse gas intensive industrial production of iron metal. The unexpected, high solubility of iron oxide in lithiated molten carbonate electrolytes was demonstrated to lead to the facile splitting of iron oxide to iron metal with the concurrent release of oxygen.

Here, we consider this unusual electrolytic splitting as a battery “charging”. We couple this with the known primary discharge of the air cathode as used in the widely studied molten carbonate fuel cell, including those using coal as a fuel, to explore the first example of a molten air rechargeable battery. In lieu of iron we also explore the alternative use of carbon and VB2 as alternative high capacity discharge redox couples for these rechargeable cells.

—Licht et al.

Illustration of the charge/discharge of the iron molten air battery in molten carbonate. The charging or discharging process is indicated by red or blue text & arrows. Licht et al. Click to enlarge.

For the carbon molten air battery, carbon formation during molten carbonate electrolysis provides the “charging”; molten carbonate cells have been widely probed as robust fuel cells. The combination of these two processes form the basis for the carbon molten air battery.

On the other hand, the researchers note, the foundation of understanding of the electrochemical VB2 molten air system “is small; there is little or no prior information pertaining to an electrochemical path for recharge of the molten vanadate (V2O5) and molten borate (B2O3) discharge products”.

The studies reported in the paper are less advanced that with the other two chemistries due to the scarcity of prior fundamental electrochemical knowledge of the molten system.

Advancing the new class of molten air batteries will require the exploration of an extensive range of parameters such as electrolyte and gas composition; electrode morphology; temperature; and cycle rates, the authors point out.

For example, while these experiments have been conducted in the 700 °C to 800 °C temperature range, the molten carbonate electrolyte has a wide range of electrolyte opportunities, such as through the use of mixed alkali carbonate eutectics which exhibit a minimum melting point below 400 °C. Enhancements of the morphology and modifications of the electrocatalytic nature of the air electrode should improve energy efficiency of the cell. A range of cell configurations with lower polarization (with similar discharge potentials, but supporting significantly higher current density) will be reported in a future study.

—Licht et al.


  • Stuart Licht, Baochen Cui, Jessica Stuart, Baohui Wang and Jason Lau (2013) Molten Air - A new, highest energy class of rechargeable batteries. Energy Environ. Sci. doi: 10.1039/C3EE42654H



Here is an Arxiv link that works for me right now:


I wish I had the knowledge to analyze this better- looks like they are targeting 400-800 C operating temperatures so obviously difficult for using with automobiles. The Carbon molten air battery has a lower volumetric capacity (19,000 Wh litre vs 27,000) versus the VB2 but much higher gravimetric density (8,900 Wh,kg versus 5,300). Hard to get a sense of the cycle life- a chart on page 9 appears to show 7 cycles on the carbon at 750C but gives little to go on.

Cost is obviously a factor for Grid Storage and this appears to have a potential for low cost.

Could be an amazing development in the end- hopefully someone more qualified can comment! Someone needs to interview the authors ASAP!


How long will it take to mass produce a practical-affordable version?


During charging, these cells produce pure oxygen as an effluent.  I'd like to see how this fire hazard gets managed.


27Kwh/liter for the third chemistry? That is definitely worth exploring.

My second-gen Prius has a ~45 liter tank. Even if we take a more practical (say 15kwh/liter) real-life capacity, that would give a 675 kwh battery. With the electric consumption of the Prius (150 Wh/km) this would result in a 4500 km range.

Now, that would be a car that crushes all of the petrol-based competition.


Even 10 kWh/liter would be great, and iron is very cheap.  450 kWh in a Prius-tank-sized package would drive a run-of-the-mill EV around 1500-1800 miles.  That pretty much eliminates "range anxiety".

As I keep saying, it only takes one such development hitting the market to kill the ICE and the oil industry.


@Engineer-Poet; Could not oxygen be fairly easily vented to the atmosphere using a diffuser and fan? I imagine that forced air blown into the area would quickly dilute the oxygen to non-hazardous levels. For a non-stationary application like automotive, those fans probably wouldn't need to be running very often.


Recharging a 450 kWh to 675 kWh battery pack may take a while.

Many new homes with 400 Amps/220 VAC distribution panel could probably spare 200 Amps and recharge those batteries at a 44 kWh rate during off peak periods and/or night time. Ten hours at 44 kWh could push about 440 kWh or enough energy for almost 3000 miles or about 2 months for the average user?

ICEV and Oil fans would go insane?


These could be combined with SOFC/MCFC to do a V2G system. Heat, cool and power your home while having transportation when you need it with no imported oil.


A battery operating at 700 degrees C is definitely going to be a challenge for mobile applications. Insulation, seals, etc. will be a nightmare.


Stationary applications might be better. The trick with SOFCs is not to thermally cycle them any more than you have to. You keep them running at temperature continuously, even at partial output.

It might still make a good CHCP system with an SOFC, the stack could be smaller then the battery would help with the high demand times. Either way this is a good bit of research.

You could compress the oxygen for later reuse, it would cut down on contaminants and extra moisture. I don't think fire danger would be a problem when properly mixed with the nitrogen in the air.

Recharging a 450 kWh to 675 kWh battery pack may take a while.

So would discharging it.  But if you just plugged the car in all the time when parked, most people wouldn't ever need to worry about running low.  If you had even a 500-mile battery, being down 200 miles of range just wouldn't be an issue for most people 99% of the time.  If you were planning to go someplace, you'd make sure the car was fully charged.  If you're just commuting, hooking up to 120 VAC 12A (level 1) for 20 hours a day would do you for almost 150 miles at 200 Wh/mile.  More likely you could have the car backfeed the grid during the day, and charge at night.  Hello nuclear baseload, bye-bye gas-fired peakers and fracking.

The Tesla Model S only carries 85 kWh, max.  A 10-liter, 100 kWh battery lets just about anything out-range the Tesla.  Recharging a 100 kWh battery at 480 V / 32 A takes under 7 hours, so if you charge overnight and during meal breaks you can run pretty much all day without worrying about running out of juice.

Could not oxygen be fairly easily vented to the atmosphere using a diffuser and fan?

Imagine oxygen building up to 30% or more in a space like a garage.  Now imagine someone walking into it with a lit cigarette, and dropping it on something flammable when it flares up in their face.  I wouldn't want to be near it.

Building codes for enclosed spaces with chargers may need to specify ventilation fans and sprinklers.  But they need this even more so for lead-acid battery banks, so the knowledge to do it right is already there.

A battery operating at 700 degrees C is definitely going to be a challenge for mobile applications.

I'm visualizing a nickel-alloy inner vacuum shell with a stainless outer and an opaque aerogel filling for insulation.  If you only have 10 or 20 liters of volume to worry about, you can afford to get fancy.


@EP great points all around

Two things I'd like to add. First, with such a high potential specific energy, you can afford designs that reduce said specific energy by 20, 30, 50+% and you still end up with an incredibly dense energy source. Second, I get the feeling that this type of battery architecture would have a practically "unlimited" cycle capacity.


We can hope, can't we?  On the other hand, unless the metal electrode is a liquid we can expect the usual spoilers like dendrite formation to raise their ugly heads.

Even that may not matter if the cells can be designed to be reconditioned easily and cheaply.  No matter how this works out, it is going to be a fantastic story about the way science works and the myriad twists and turns and dead ends on the road to success.

Roger Pham

>>>>(i)"Recharging a 450 kWh to 675 kWh battery pack may take a while."(/i)

Why bother fast-charging a 450-kWh BEV? Why bother having a BEV with such a stupendous capacity at all?
An optimized, clean-sheet-design PHEV with a small 2-3-cylinder ICE of about 1 liter displacement, coupled with a small but reasonably-high-power-and-energy-density battery pack will be much more advantageous than a pure BEV having 100-450-kWh of super-energy-dense battery pack.

This is because a pure BEV will need fast-charging infrastructure to be built everywhere, yet is not often used, hence a very poor business model for this to takeoff. A PHEV takes advantage of existing fast energy fill-up at all existing gasoline stations, thus no new infrastructure investment.

Furthermore, for long trips through desolated areas, a PHEV can carry extra fuel in external tanks to cover distances as far as thousands of miles. A pure BEV will have problem with that.

A pure BEV will suffer from calendar life degradation of the battery pack before its cycle life can be used up. A PHEV can have a much smaller battery pack replaced several times during the life span of the car, thus is able to max out the battery's cycle life without significant calendar life degradation.

In another words, existing battery technology of 200Wh/kg Lithium battery is just fine and dandy for a very practical PHEV that can compete with ICEV in all aspects. Rapid adaptation of this existing battery technology to large-volume production will be far more practical than continue to wait for a silver-bullet battery solution like Li-air or solid-state battery...etc.

Existing engine and battery technology is sufficient to create a PHEV that is price competitive with existing ICEV as well as having comparable internal space and curb weight. The OEM's simply do not yet want to build such a PHEV!!!


When someone brings a sufficiently cheap, efficient and long-lived battery to market, it will be snapped up by the utilities to time-shift generation from overnight to peak demand times and provide regulation and spinning reserve.  The storage will be distributed at sites like substations.

So long as all that storage is just sitting there, utilities are going to look to other revenue opportunities.  Fast charging is one of them.  If you've got a battery waiting to supply 3 megawatts in the event of a breaker tripping, feeding 300 kW for a couple of fast chargers is nothing.

But that's the future.  Today's "fast charge" is petroleum, and this will be the case until probably 2020.  In the mean time electricity can chew away at petroleum consumption by powering the first 20-30 miles of each day's driving, which is the bulk of it.


I seem to recall that a V2O5 route was listed way back in the 60's as one of a handful of promising recyclable quasi-catalytic routes of producing hydrogen, as researched by Italian scientists. This was apparently a purely theoretical exercise as it was one of about 400 or so routes conjectured (You'd have to look it up in an old energy tech anthology edited by David Pimentel, which I do not have). It may be that something of this hot cell technology can be utilized to do just that. If so, the dangerous part of heating the process can be left to a protected central location, and you could just draw of the hydrogen for your cars instead. Water instead of air would be the source of oxygen and hydrogen.

This is similar to an iron smelting system reported on this blog that utilizes lithium carbonate and CO2, which has a lower free energy barrier than sci texts previously reported. Seems that the new Iron Age is upon us.

Roger Pham

Good point, E-P. I didn't think of the idea of battery-based spinning reserve for electric utility that can double as a BEV ultra-fast charger.

However, by the same token, an H2-generating electrolyzer and FC system can be considered as a flow battery that can handle seasonal-scale of energy storage. The efficiency is not as good as Lithium-based batteries, but perhaps the efficiency can be improved with time, or the waste heat can be used for other purposes to improve the round-trip efficiency. The energy density and power density of H2 storage and FC already exceeds the best of current Lithium chemistries by a large margin. A local H2 piping system can serve both as H2 storage as well as H2 dispenser for stationary FC for cogeneration CHP for grid electricity and space heating and water heating, as well as dispenser for FCV's that will be commercially available by 2015.


or the waste heat can be used for other purposes to improve the round-trip efficiency.

Our city has a number of district heating systems that put solar and industrial waste heat to use. I can see the waste heat from H2 electrolyzers and FC systems used likewise IF the temperatures are high enough.

Kit P

“Our city has a number of district heating systems that put solar and ”

The socialist government of British Columbia provides ai vin his welfare check. District heating implies winter. Winter in BC has no solar resources.

I am not surprised anymore by how clueless about the natural world those who claim to care about the environment.


Speaking of chewing away, my vehicle's lifetime average fuel economy surpassed 150 MPG today.  (It was close to that 3 weeks ago, but a 300-odd mile trip bumped it down to the low 130's and it's taken this much daily running on juice alone to run it back up.)

Still haven't put fuel in the car since June.



Aren't you describing the PHEV Chevy Volt, the PHEV Prius, and PHEV Ford Focus?

I have doubts whether a 2 or 3 cylinder engine would provide the power density necessary to push a family sedan around, but I agree with you that a PHEV is a good solution for covering both short trips and long trips using the cheapest, most available fuel for the application.


Roger Pham


Thanks for discussing this issue. Yes, the PHEV Ford Focus is renamed as C-Max Energi. It has a 4-cylinder-2-liter-144-hp engine. A 144-hp engine is desirable for the HEV version because the battery is rather weak. However, a PHEV has much more powerful battery to supplement the engine, so that only 72 hp, or 2 cylinders, would be desirable. Remember that the Prius gen II HEV has only ~72 hp from the 1.5-liter 1NZ-XFE engine and only 30 hp of battery power for supplemental power, and the Prius gen II has plenty of power for acceleration. The Prius gen III has around 33% more powerful engine and battery, yet does not feel any faster in acceleration when I drove both of them. Perhaps I used only a fraction of the power available for acceleration and therefore cannot appreciate the power different between the Prius gen II vs. the Prius gen III, yet I can keep up just fine with traffics in either version. This goes to show that hauling around dead weight of engine, drive train, motors, and battery is not a good idea for efficiency, nor for purchasing cost reduction.

The higher costs of HEV's and PHEV's have been barriers for more market penetration of those vehicles. The reduction in cargo space is another turn-off for PHEV's. By reducing the engine to 2 cylinder up front, some battery can be placed up front to reduce the loss of cargo space taken up by the battery pack behind the rear seat. Cost reduction is realized with downsized and simplified engine and motors. Furthermore, the fuel tank of the PHEV and HEV can be downsized to provide a range of only 300 miles instead of a range of 500-600 miles, thereby allowing more luggage space. The weight reduction from all of these can allow further weight reduction from the suspension and drive train and body and chassis in a clean sheet design, allowing a PHEV to have comparable weight and handling to a similar ICEV.

Please remember that the target audience for HEV and PHEV are not sport car enthusiasts desiring neck snapping acceleration, but efficiency-minded people who prefer to get around in a more gentle fashion.

Roger Pham

@ai vin,
High-temperature PEM FC is available as introduced by VW in order to reduce cooling needs, and the temperature is high enough for steam generation for industrial use. A home hot water heater and space heater require a lot lower temperature, perhaps 60-70 degrees C instead of above 100 degrees C.

The comments to this entry are closed.