DOE announces marine and hydrokinetic open data effort
CO2 Solutions’ enzymatic carbon capture technology performance exceeds targets in oil sands project

Hyundai Motor researchers report improved Li-sulfur battery performance with new sulfone-based electrolyte

Researchers from Hyundai Motor have found that the use of a new sulfone-based electrolyte greatly improved the capacity and reversible capacity retention of a Li-sulfur battery compared to the performance of ether-based electrolytes. In a paper presented at the SAE 2104 World Congress in Detroit, they reported that use of the sulfone-based electrolyte increased capacity by 52.1% to 715 mAh/g and capacity retention by 63.1% to 72.6%.

Lithium-sulfur systems are of great interest as a “beyond Li-ion” solution with increased energy densities that would enable much greater electric vehicle range. The Li/S system has a high theoretical specific energy of 2600 Wh kg-1; however, rapid fading of charge capacity is a well-known issue (e.g., earlier post). The poor long-term performance has been associated with both the shuttling of polysulfides dissolved into the electrolyte, in addition to irreversible deposition of solid lithium sulfide (Li2S) and other mixtures of insoluble discharge products on the cathode.

Chemical processes in Li/S battery include Li ion dissolution from Li metal anode and sulfur reduction to Li polysulfides (PS, the series of sulfur reduction intermediates) on the sulfur cathode while discharging (S8→Li2S8→Li2S6→Li2S4→Li2S), and reversible chemical reactions occur during charging. Among the PS formed in this mechanism, Li2S6 and Li2S4 are soluble in electrolytes. PS solubility plays an important role to improve cyclability as increasing the sulfur utilization.

Ether type solvent has been considered as a suitable electrolyte for Li/S battery because it has good PS solubility and chemical stability. Meanwhile, dissolved PS causes redox shuttle resulting in a low COlumbic efficiency, poor cycle life and self-discharge. Therefore, this work aimed at developing a new electrolyte in order to prevent redox shuttle and improve cyclability.

—Shin et al.

In their study, the Hyundai researchers compared the performance of 5 single component ether-based systems (DME, DEGDME, Triglyme, TEGDME and DIOX); one binary system (TEGDME:DIOX); and three versions of a ternary system with sulfone (TEGDME:DIOX:Sulfolane at ratios of 1:1:1, 1:1:2, and 1:1:3).

They assembled coin cells for electrochemical testing using a sulfur cathode and Li metal foil as an anode, with a polyethylene separator between them. Cycling test were performed between 1.5V and 2.65V and room temperature at C/20 rate.

Among the single component ether systems, use of DME resulted in the highest capacity of 878 mAh g-1, with DEGDME a close second at 857 mAh g-1. However, the cells with these electrolytes showed drastic capacity fade after 6 and 2 cycles, respectively.

While cyclic ether, DIOX, showed 1,040 mAh g-1 at first cycle, this dropped to 640 mAh g-1 at 12 cycles. The high initial discharge capacity showed that DIOX appeared to be effective on developing high capacity; however, after 13 cycles it evidenced a drastic capacity decrease.

TEGDME showed a low initial capacity of 200 mAh g-1, but it did not show drastic capacity fade.

The researchers combined TEGDME and DIOX into a 1:1 binary system to investigate the synergy of the good cyclability of TEGDME and the high capacity of DIOX.

The cell with the binary ether electrolyte showed first discharge capacity of 1057 mAh g-1 and 470 mAh g-1 after 20 cycles. Compared to the single component results, this cell showed good cyclability. However, the issues large drop in capacity after the first cycle and the low reversible capacity retention of 44.5% after 20 cycles remained.

The researchers then inserted a glass filter between the electrodes to restrain the high resistance around the electrodes in the cell with the binary electrolyte. (The glass filter absorbs electrolyte, thereby preventing a deficiency of electrolyte next to the electrode.) This served to increase overall capacity to 605 mAh g-1 after 20 cycles, lowered the capacity decrease after first cycle, and improved capacity retention.

Chemical analysis suggested that sulfone solvent could form a protective layer on the anode surface, and prevent the PS shuttle by blocking the reaction between the Li anodes and PS. In addition, the protective layer can mitigate the crack formation on the surface observed with the other electrolyte systems, the researchers determined.

The Hyundai team used Sulfolane (a sulfone-based solvent), as it is already known as a suitable Li battery electrolyte. They prepared three ternary compositions of electrolyte, adding different amounts of sulfolane to the binary TEGDME : DIOX mixture.

They found that the 1:1:2 (TEGDME:DIOX:Sulfolane) mixture (TDS2) showed the best cyclability, as noted above with capacity of 715 mAh g-1. TDS1 also showed improved capacity and retention: 674 mAh g-1 and 68%. Cycle performance worsened in TDS3.

The researchers also found that cracks on the anode surface diminished significantly.


  • Shin, N., Ryu, K., Kim, Y., and Lee, H. (2014) “Improved Cyclic Performances of Li-Sulfur Batteries with Sulfone-Based Electrolyte,” SAE Technical Paper 2014-01-1844 doi: 10.4271/2014-01-1844



You might have to build hybrid battery packs with LiS for main energy storage, LiIon for power, and possibly Supercaps for recovering braking energy suddenly.

It sounds complicated, but once you got it going, you could have a viable electric car, without a ton of LiIon batteries.


One more possible solution for highway capable EV by 2018-2020...

I said it first ! HarveyD are you loosing heart ???



You are on to something, the range extender idea gets away from one battery pack having to do everything. Power dense ion batteries and energy dense sulfur batteries would get us a LEAF with 200 mile range at a lower price.


Hope that Treehugger is right. I'll need a new afordable extended range BEV in that time frame.


Is there any technical reason preventing supercaps being integrated with lithium-based batteries, as the CSIRO did with lead-acid + supercaps in the Ultrabattery?


Here we go again. The first paper announcing button-cell results and optimists are projecting a 5-7 year adoption for global BEV application. You have yet to see any replication of this work -- add another year or two. Then a cell of practical size. Another year. Then a representative battery -- one or two more years. Follow that with process industrialization, environmental and safety testing, development of compatible BMS/thermal mgt, toxicological studies and approvals.... the vehicle market is a decade away if (IF!) the chemistries can even work safely on a larger scale.

There's nothing wrong with dreaming and hoping and prodding political agencies for R&D attention. But comments columns like this generally go from speculation to unfounded prediction to dark prognostications of patent or import barriers that will hold it all back. If there is a production BEV battery with twice the currently available specific energy, half the current cost, QC capability of 80%/30min, and no significantly adverse weather-related behaviors by MY2018, it will be a major accomplishment, and we should all cheer such an achievement.

5x improvement in any of those metrics? No way.


Indeed this seems more like a step on the voyage to more knowledge, than a concept for a practical battery in a few years time.

But conventional Li-ion batteries have still much room for improvement and capacity will slowly and steadily grow until LiS batteries or another advanced chemistry are ready for mass production. That first generation of suboptimal LiS batteries will only be marginally better than the Li-ion batteries at that time.

From the perspective of the customer there will never be a breakthrough, just steady, evolutionary improvements. And that's all we need. Growth of EV sales is hampered by more things than limited battery capacity. As long as they can keep 7% improvement per year going, that's fine by me.


I don't know if this method will bring us higher capacity per weight and/or volume, but the idea of having two packs makes sense. We might end up with a 40 kWh pack the same size and weight as a 20 kWh pack, half the cost but less power density.

The 40 kWh pack can produce 80 kW max. but the 10 kWh lithium ion can deliver 100 kW if necessary. The larger pack recharges the smaller pack and the smaller pack can produce acceleration current and take in braking energy. The larger pack may not quick charge, but it costs half as much while giving you twice the range.

This is where marketing comes in, find out what is acceptable to the buyer and you know more about the goals. If most people don't need quick charge because they charge at home but want a 200 mile range at lower cost, then that is a worth while goal.


Alternatively we could do something we know perfectly well how to do, with far greater energy density than any batteries in prospect, and put in a fuel cell system with a 12kwh battery pack so that it is a plug in hybrid.

Batteries aren't going to hit the ~1500Wh/kg that a fuel cell system including the carbon fibre storage tank can hit anytime soon.


Again, marketing....find out if people want to sit on 10,000 PSI hydrogen tanks.


What on earth are you talking about?
The safety of the carbon fibre tanks is well established and thoroughly tested, with vehicles having done millions of miles and having had countless fills.
Hydrogen vehicles are no more dangerous than petrol cars.
You are simply spreading FUD.


If few will buy it, it will do less good. You can tell prospective car buyers it is perfectly safe, but will they believe you? Will they make the choice to purchase when there are alternatives that they believe are safer? I am not making a judgment on hydrogen tanks, just pointing out buyer perception as a factor.


I'll make a judgement on Hydrogen tanks. They leak. H2 molecules are so small that it's very difficult to completely seal them in. What happens over decades if just 2% of all Hydrogen fuel used eventually ends up in the upper atmosphere? Oh, and the word "Hindenberg" comes too readily to people's minds. Mostly, I just don't like the conversion losses with electrolysis, and reforming methane is still releasing greenhouse gases into the atmosphere, so that's a bait-and-switch. Hydrogen is a storage medium, not an energy source, and compared to near future batteries, H2 is probably a disadvantageous path to go down.


Biff, cost, cost is the issue with supercaps. Supercaps have very little energy, but high rates and thus are good for braking energy recovery. However, low cost of materials and manufacturing needs to happen.


With the right chemistry, super caps may not be needed. Toshiba and Altair make titanate batteries that can take a huge charge in a short time. 10-12 kWh of cobalt or manganese cathodes can take a lot of brake energy.

We discussed the idea of a battery range extender using two packs on Green Car back in 2008. Not withstanding any attempts by Tesla to patent the idea, it stands as a good concept known to many. Musk says he does not like patents, but he seems to encourage his people to do so.

The comments to this entry are closed.