Li-S Battery Shows Cycle Performance Comparable To Li-Ion, Double The Energy Density

APR 25 2015 BY MARK KANE 23

Good news is coming from the battery front where researchers in South Korea and Italy announced a milestone for lithium–sulfur cells.

Their new Li-S cells, with energy density over two times higher than lithium-ion (497 Wh/kg), ran through 500 cycles with capacity remaining at about 85% (∼750 mAh g–1).

We believe that the cycles themselves were low-current – there is a value C/3 (3 hours charging/discharging) mentioned in the article in the part about cathodes, which still would be not enough for a car, but efficiency is brilliant.

Work is presented in the ACS journal Nano Letters:

“Lithium–sulfur batteries could become an excellent alternative to replace the currently used lithium-ion batteries due to their higher energy density and lower production cost; however, commercialization of lithium–sulfur batteries has so far been limited due to the cyclability problems associated with both the sulfur cathode and the lithium–metal anode. Herein, we demonstrate a highly reliable lithium–sulfur battery showing cycle performance comparable to that of lithium-ion batteries; our design uses a highly reversible dual-type sulfur cathode (solid sulfur electrode and polysulfide catholyte) and a lithiated Si/SiOx nanosphere anode. Our lithium–sulfur cell shows superior battery performance in terms of high specific capacity, excellent charge–discharge efficiency, and remarkable cycle life, delivering a specific capacity of ∼750 mAh g–1 over 500 cycles (85% of the initial capacity). These promising behaviors may arise from a synergistic effect of the enhanced electrochemical performance of the newly designed anode and the optimized layout of the cathode.”

Source: Nano Letters via Green Car Congress

Categories: Battery Tech

Tags: , ,

Leave a Reply

23 Comments on "Li-S Battery Shows Cycle Performance Comparable To Li-Ion, Double The Energy Density"

newest oldest most voted

500 cycles is just barely adequate for an EV unless you are talking about a vehicle with 200-300 miles of range like a Tesla. The reason being you won’t cycle the batteries as much. Not sure this would be any good for a PHEV.

True, but… A 240 mile EV would be at 120k after 500 cycles. The choice of battery for a manufacturer would come down to a 70kWh pack of Li-S or a 35kWh NMC for the same price.
They’d likely have similar performance. The Li-S would be listed as 240-300 miles of range, while the NMC would be 120-150. The NMC would last for 240k miles or more, The Li-S probably a little over 100k.
So the Li-S would be easier to sell and last just long enough to cover the standard 100k mile drivetrain warranty. Seems like an easy decision as a manufacturer.

Both chemistries could be combined in the same pack, with the Li-S being a kind of range extender.
You could have (say) 60% of the capacity being NMC and the rest Li-S.
If we’re talking a total 200mi battery, the NMC cells would be discharged first, with a 120mi range, so in normal daily usage or even short intercity trips you’d never access the Li-S part.
The Li-S would be used only on longer trips, so would last much longer.

This type of dual-chemistry scheme is considered the best way to use Alu-Air bstteries as well, see:

I believe Tesla had a patent for a hybrid battery pack. Not sure how they got the patent actually as it seems an obvious idea. It will be interesting if the Model 3 uses this approach but I guess they’ll probably stick with what they has worked for them so far.

These test cycles are deep cycles,
from 100% down to 0%. As we all know, the most damage is done at the ends of the charge state (near 100% and near 0%). That is why manufacturers tune their bms’s so that they do not use the full capacity. Supposedly, my Zoe has a 27 kWh battery of which on;y 23 are usable. The rest is a buffer to improve longevity.

If you would do the same with these lithium sulfur batteries, they would likely last as long as the current car batteries.

That doesn’t change the fact that all batteries are tested this way, including the batteries that are in current EVs and they get a rating of thousands of cycles. All except Tesla who also has around 500-600 cycles. But as I mentioned, they get away with it because of the size of their battery pack and so it takes much longer to put that many cycles on it.

I don’t think li-ion batteries are typically tested for longevity/ capacity loss by repeatedly draining them to 0% charge while cycling. That causes li-ion batteries to age prematurely.

David, see Mint’s comment below that cites the paper.

Do you have a source that typical lithium ion batteries last thousands of full DoD cycles? I don’t think these batteries exist. As far as my general knowledge of lithium batteries goes, they only last thousands of cycles if you limit DoD, which the BMS does in all EV’s currently on the road.

arne-nl said:

“These test cycles are deep cycles, from 100% down to 0%.”

May I ask what your source of info is for that statement? Is the info in the original paper? It’s behind a paywall so I can’t read it.

One would not expect typical lithium ion batteries to be tested for longevity/cycling by being repeatedly drained to 0%, as that causes premature aging. Typical deep cycling in li-ion batteries uses 80% DOD (Depth Of Discharge), which could mean, for example, charging to 90% and draining to 10%.

Of course, lithium-sulfur batteries may have different characteristics, and perhaps are not damaged by being repeatedly drained to 0%.

In the paper, the caption under the cycling graph is:

“Figure 4. (a) Voltage profiles of the full cell adopting the lithiated Si-SiOx nanosphere/DME/DOL (1:1 v/v), Li2S8 (0.05 M), LiTFSI (1 M), LiNO3 (0.4 M)/AC-S structure and including a GDL current collector, cycled at 0.1C (black curve), 0.2C (red curve), and 0.5C (green curve) rates. (b) Voltage profiles and (c) cyclic responses of the lithiated Si-SiOx/DME/DOL (1:1 v/v), Li2S8 (0.05 M), LiTFSI (1 M), LiNO3 (0.4 M)/AC-S full cell cycled at 1C rate. The upper and lower voltage limits are, respectively, 2.8 and 0.8 V. 1C = 1675 mAh g–1 versus overall sulfur weight; temperature, 30 °C. Inset: magnification of Coulombic efficiency.”

The key point is cycling between 2.8V and 0.8V. That’s full DoD.

But even if it wasn’t, your point would be moot. A slightly smaller cycling range wouldn’t take away from the importance of this work at all.

Your statement below of 15% loss after 500 cycles being insufficient for EVs isn’t valid at all, either.

Wonderful to see progress being made in alternatives to lithium-ion. It’s true that 15% loss of capacity in only 500 cycles isn’t sufficient for EVs, but hopefully this can be further improved.

Let’s remember than the 1st generation GM EV1 was powered by lead-acid batteries. Since then, we’ve seen EVs powered by NiMH, lithium ion, and lithium ion polymer batteries. Ni-cad batteries were even used in some niche applications. Li-ion polymer isn’t the end of battery development; it’s just the current state of the art.

I look forward to the next fundamental improvement in EV battery tech, whether that’s lithium-sulfur, solid state batteries, metal-air “batteries” (actually fuel cells), flow batteries, the “all-electron battery”, or something entirely unexpected.

See my comment to David Murray. In real life, the batteries are not allowed to cycle fully, and they will last far more than 500 cycles.

Well, I’m somewhat skeptical of your claim. I’d like to know what your source of info is, regarding tests done on these batteries.

“Deep cycle” doesn’t mean charging to 100% and discharging to 0%. Not for deep cycle lead-acid batteries, and not for lithium-ion batteries either. Typical li-ion deep cycling is an 80% DOD (Depth Of Discharge) cycle, avoiding charging to 100% or draining to 0%.

He is correct. Tesla initially guaranteed their first investors specifically that their battery pack would retain 70% capacity after 1,000 100%-0% charges and discharges. It does cause much greater wear than partial discharges but reliably so, that way you can extrapolate the results to hw much the lifespan would be increased with gentler use.

At two times the energy density a Model S with 250 miles would have 500 miles. 500 miles times 500 cycles that is 250000 miles and still running at 85%.
This must all be checked and tested but that seems like at least a possible way for a 500 miles model S and 250000 miles.
The idea of an hybrid battery is also an interesting.

While we think of Cars like a Double Range Tesla for 500 miles, This could also help cars like the 62 mile iMiEV – to make it over 100 miles! Or – the Current Soul EV’s 93 miles – could easily bump up to 150+ Miles! Consider that Elon Says – ‘200 miles is the minimum for good EV’s’ – and the battery weight change of this vs. Li-Ion Cells – could help to get the pack weight down at least 40% (+/- depending on Power Capabilities) Another first step – would be – if this is very stable chemistry – to put it out to the EV Conversion Market Leaders as a Alpha Test for the field! (Like – and others!) Even my Firefly – ‘Electricfly’ from Lead Acid to this would be a massive upgrade, and even about almost 5X as good as the LiFePO4 Headway cells I have been considering which are at about 105 Wh/Kg!! (Which Themselves are about 3X as good as Lead Acid Batteries!!) Imagine a EV1 – re-built and using the same space/Weight – in this chemistry – going from Lead Acid with about 35 Wh/Kg to 497 Wh/Kg! Over 14X the… Read more »

Actually, it wouldn’t take future tech like Li-S to upgrade the i-MiEV’s range to 100+ miles… The battery design was finalized in 2009, and is 80Wh/kg…

Simply using the Soul’s modern 200Wh/kg cells would easily get you a 100mi i-MiEV.

“Their new Li-S cells, with energy density over two times higher than lithium-ion (497 Wh/kg”

It’s curious that they mention the WH/Kg part of energy density but don’t mention WH/L. What kind of volume do they get for a similar weight to Li-ion? This could be a problem in the extreme if say they get double the WH/Kg but that advantage becomes somewhat negated if it has triple or quadruple the volume.

I assume they just are simplifying by saying double the density and that it is assumed that we all know that it also has at least double the WH/L to that of Li-ion. If not it would be slightly dishonest to say double the density if the volume didn’t jibe.

Wh/L is probably similar to other Li-ion. The Wh/kg advantage comes from the fact that your replacing heavy (and expensive) metals like nickel, cobalt, and manganese with light and very cheap sulfur.

That is unfortunate, as Teslas are almost already too big, nearly venturing into Flavor Flav territory. Still, the reduction in weight is welcome.

I would absolutely -not- assume the volumetric energy density (size) would be decreased proportionally with the gravimetric energy density (weight).

Sadly, far too often these claimed “breakthroughs” are just optimizing battery chemistry to maximize one or two characteristics at the expense of making the others far worse than a more balanced chemistry.

Now, as I understand it, the -potential- is there for significant improvements in energy density with lithium-sulfur; both volumetric and gravimetric. But any claim such as this needs to be tested by a third party before we can consider it reliable. In the battery industry, unfortunately, claims for improved tech are seldom wholly true.

“My my top advice really for anyone who says they’ve got some breakthrough technology is please send us a sample cell. Okay? Don’t send us PowerPoint. Just send one cell that works with all appropriate caveats. That would be great. That sorts out the nonsense and the claims that aren’t actually true. Talk is super cheap. The battery industry has to have more B.S. in it than any industry I’ve ever encountered. It’s insane.” — Elon Musk

What is also noteworthy is that the KWh/Kg is the dominant factor in everything that fly, from airplane to helicopter. Like the Lithium ion battery was key to electric cars, the Lithium Sulfur battery could be the key to the skies. We could go from a two seat electric training plane to an electric version of an ATR42, which would be a nice step all the way to an electrofan 777.


Anyone can cite capacity loss over 500 cycles for current gen LiOn batteries with same discharge/charge characteristics?

Comparison would be in place here.