UC Berkeley/Berkeley Lab teams develops high-rate, long-life Li-S battery with Li2S-graphene cathode
01 May 2015
|
Li2S/GO@C Nanosphere. Credit: ACS, Hwa et al. Click to enlarge. |
Researchers with appointments at both UC Berkeley and Lawrence Berkeley National Laboratory have developed a high-rate and long-life Li-sulfur battery cell. The cathode material is a core–shell nanostructure comprising Li2S nanospheres with an embedded graphene oxide (GO) sheet as a core material and a conformal carbon layer as a shell.
The Li2S/GO@C cathode exhibits a high initial discharge capacity of 650 mA·h g–1 of Li2S (corresponding to the 942 mA·h g–1 of S) and very low capacity decay rate of only 0.046% per cycle with a high Coulombic efficiency of up to 99.7% for 1500 cycles when cycled at the 2 C discharge rate. A paper on their work is published in the ACS journal Nano Letters.
The Li/S cell has attracted great attention because of the need of the electrical vehicle (EV) market for high specific energy batteries (∼350 W·h kg−1 at C/3 discharge rate), which greatly exceeds the practical specific energy of current Li ion cells (100−200 W·h kg−1). However, despite the great advantages of Li/S cells, the early S-based cathodes in organic electrolytes showed a low utilization and a poor cycle life owing to several major problems: (i) The insulating nature of Li2S and S that are the final products of the S electrode at the fully discharged and charged states, respectively. (ii) A large volume change of the S particles during cycling (∼80%) resulting in mechanical degradation of the electrode. (iii) Highly soluble intermediate species (polysulfides, Li2Sn, n = 4−8) in most organic liquid electrolytes, which causes the loss of active material from the cathode and the polysulfide shuttle effect. When the polysulfides are dissolved into the liquid electrolyte, they can diffuse back and forth between electrodes and can form insoluble Li2S (or Li2S2) on the surface of the Li metal electrode, which leads to lower Coulombic efficiency.
—Hwa et al.
Key factors to improve the electrochemical performance of Li/S cells are to increase the electronic conductivity of the cathode and to suppress the polysulfide dissolution as well as the mechanical stress caused by the volume change during cycling. Numerous approaches have been proposed: nanofabrication of S (or S-based composites); chemically (or mechanically) protective materials on the S particles; and composites with mesoporous carbon or graphene oxide (GO) that can act as S immobilizers.
GO is very attractive for stabilizing the cycling performance of S-based cathodes because the reactive functional groups on the surface of GO can form bonds with S, indicating that S (or polysulfides) can be captured by those functional groups, the Berkeley team noted.
Recent work has begun exploring the use of lithium sulfide (Li2S, theoretical specific capacity: 1166 mA·h g−1) as the initial cathode material instead of S (e.g., earlier post, earlier post). With Li2S as the cathode material, the mechanical damage of the cathode due to the volume expansion of S particles (up to 80%) caused by the lithiation process during discharge can be reduced because Li2S particles already occupy the maximum volume relative to S.
In addition, the prelithiated state of the Li2S cathode can be coupled with Li-free anodes such as silicon (Si) and tin (Sn), thereby avoiding the problems currently associated with Li-metal anodes such as dendritic growth.
However Li2S suffers from very poor electronic conductivity, polysulfide dissolution and the shuttle effect, which cause low S utilization, low Coulombic efficiency, and rapid degradation during cycling.
To address those issues, the Berkeley researchers used Li2S/GO nanospheres with a conformal carbon coating on the surface (Li2S/GO@C). Their material offers a number of benefits:
The conformal carbon coating not only prohibits polysulfide dissolution into the electrolyte by preventing direct contact between Li2S and the liquid electrolyte, but also acts as an electrical pathway resulting in the reduction of the electrode resistance.
The spherical shape of the submicron size particles can provide a short solid-state Li diffusion pathway and better structural stability of the carbon shell during cycling.
Void space is created within the carbon shell during charge, and it will provide enough space to accommodate the volume expansion of up to 80% during discharge. As a result, better structural stability of the carbon shell can be secured because the carbon shell will not need to expand during cycling.
Even if some percentage of the carbon shells is broken due to physical imperfections, the GO in the particles can act as a second inhibitor for polysulfide dissolution due to its S immobilizing nature.
|
Cycling performance of the electrodes cycled at various rates. Credit: ACS, Hwa et al. Click to enlarge. |
The Li2S/GO@C nanosphere cathode demonstrated promising electrochemical performance:
Prolonged cycle life (1500 cycles) at the 2.0 C discharge rate (1.0 C = 1.163 A g−1 of Li2S) with a high initial capacity of 650 mA·h g−1 of Li2S (corresponding to 942 mA·h g−1 of S) and 699 mA·h g−1 of Li2S (1012 mA·h g−1 of S) at 0.05 C after 400 cycles at 2.0 C discharge; and
excellent capacity retention of more than 84% with a high Coulombic efficiency of up to 99.7% after 150 cycles at various discharge C-rates (2.0, 3.0, 4.0, and 6.0 C discharge rates).
Resources
Yoon Hwa, Juan Zhao, and Elton J. Cairns (2015) “Lithium Sulfide (Li2S)/Graphene Oxide Nanospheres with Conformal Carbon Coating as a High-Rate, Long-Life Cathode for Li/S Cells” Nano Letters doi: 10.1021/acs.nanolett.5b00820
If affordable when mass produced, it could be OK for 500 Km BEVs?
Posted by: HarveyD | 01 May 2015 at 11:15 AM
Harvey,
I don't want 500 mile range! I don't want to pay for it, I don't want the extra weight and I don't need it. 125 miles is fine for many people and 250 miles works great for 99% of people.
I think we do a disservice to keep hoping for wildly high ranges for EVs.
Posted by: DaveD | 01 May 2015 at 12:38 PM
1500 cycles when cycled at the 2C
..good progress
Posted by: SJC | 01 May 2015 at 03:42 PM
70 mile range at 55 mph for the current Leaf battery that weighs 600 lbs ain't so hottra; but 140 miles for the same weight battery would renew my faith in Nissan, especially if they offer an reasonably priced upgrade for the older cars. In effect that means an energy density of about 300 kWH/Kg, twice the current density.
Posted by: Lad | 01 May 2015 at 06:21 PM
In fairness, Harvey suggested 500 km, 310 miles.
I agree that 150-200 would be a "no compromise" in- town car (only charge overnight at home).
But 300 miles without reducing passenger or cargo space, at a reasonable premium would also likely be a strong seller if the TCO value proposition was pitched correctly. Tesla seems to have done it effectively.
Posted by: electric-car-insider.com | 01 May 2015 at 06:28 PM
eci,
Thanks! I didn't notice he had said 500km. My apologies Harvey :)
Posted by: DaveD | 01 May 2015 at 06:32 PM
Por lo que se ve este catodo esta listo para entrar en producción 1.500 ciclos a 2C es más que aceptable para ponerlo bajo comercialización YA. La cuestión es que tienen que combinarlo con un buen anodo y el correspondiente electrolito esta hecho 1/3 del trabajo.Pero a medio plazo 2018-2020 vamos a ver si o si celdas de li-s con una energia especifica de 500-600wh/kg y EV como el Leaf por ejemplo con autonomias reales de 300-350km ,200-225 miles. Good morning from Barcelona, Spain,Europe.
Posted by: Centurion | 02 May 2015 at 04:10 AM
@DaveD.
It dépends on what you want an EV to do.
1. As a second (city car) to be recharged home every night, I agree with you that 200 to 300 Km would do the job in warm places but not in cold climate and heavy frequent snow falls like we have.
2. As a unique family car, an AWD 500Km (300 miles) BEV would be a minimum to deal with our harsh winters. The Tesla 85DF barely qualifies.
Recharging 100 kWh batteries with very low cost clean (hydro 95%-Wind 5%) electricity would not be a problem in our area. Major charging stations should get energy at about $0.03/kWh (industrials rates) on a 24/7 basis. Recharging home would cost about $0.065/kWh on a 24/7 basis anywhere on the grid.
It is just a matter of time before extended range BEVs are equipped with lighter, lower cost 100+ kWh battery pack. Installed under the floor (à la Tesla) passengers space would not be compromised.
Posted by: HarveyD | 02 May 2015 at 08:40 AM
These appears to be painstakingly created golden samples. Once these components can be created through inexpensive bulk processes then they should be ready for commercialization.
Posted by: TexasDesert | 02 May 2015 at 12:10 PM
@ DaveD
125 miles range fine if you live on an island. In the American West, you need more.
Posted by: Larzen | 02 May 2015 at 02:09 PM
@Larzen,
No, if you live anywhere around a big city, a vast majority of those people don't drive more than 30-40 miles per day. It's WELLLLLLL documented that ~80% of all daily driving averages less than 40 miles per day. Sorry, those are simply the numbers.
An option that works for 80% of people should be available. Sorry if it doesn't fit your personal situation, but that doesn't affect me and people like me.
Posted by: DaveD | 03 May 2015 at 10:46 AM
Our experience: we had a Leaf for a couple of years, and with its effective 80 mile range we were regularly looking at the range guessometer while traveling around the Bay Area. With our Toyota RAV4 EV with effective 125+ miles range (we have driven this several times, it is quite real), we simply don't look at the guessometer and the anxiety related to range became nonexistent. Now I'm looking forward to getting the 150 mile range Chevy Bolt: I like that Chevy seems genuinely interested in EVs, as opposed to Toyota's disdain for its own excellent product.
Posted by: jzj | 04 May 2015 at 04:16 PM
Over the past several years we have had numerous lab reports of the excellent energy storage, life-span, and efficiency of graphene-based lithium or sulfur or lithium-sulfur batteries, using graphene for structural integrity. Now, it is time for this problem to be taken out of the realm of a lab chemists and put into the hands of applied-science engineers for mass manufacture -- what are the big players in this space waiting for?
Posted by: jzj | 04 May 2015 at 04:27 PM
@jzj
Graphene oxide won't be cheap enough until around 2022.
If what you're waiting for is graphene based, you're going to need to wait.
Posted by: NewtonPulsifer | 04 May 2015 at 09:19 PM
jzj
There are posts from LEAF owners about the 80 mile range becoming 70, 60 and so on. No one asks what the range reduction is with Tesla after 100,00 miles.
Posted by: SJC | 04 May 2015 at 09:59 PM
Good point SJC.
When you start with 2x the range you need, it doesn't matter as much if you lose 10-20% over 10 years. Of course that comes at a cost, but the early adopters are willing to pay it so that there can be a gen 2.
The Tesla, like most EVs, has a liquid coolant temperature stabilized battery. On cold days you can set the charger by on-board timer to give it the last 5% charge before you leave home, pre-warming the battery. It stays warm by discharging during travel.
People point out the limitations of the Leaf, but it's one of the few EVs that have a heating/cooling or capacity retention problem. Nissans says that's been remedied by improved chemistry, we'll all know in a few years.
Posted by: electric-car-insider.com | 05 May 2015 at 06:31 AM
Tesla may be better, longer initial range and thermal control help.
Posted by: SJC | 05 May 2015 at 08:06 AM
@D,
No, I said a car with a real world range (that means in hot and cold weather and driving at freeway speeds) of 125 miles. I didn't say anything about 70 mile range. If that works for someone then good for them. But it's not me.
My point is that there are plenty of people who would be perfectly happy with a 125 mile "real world" range. So why shouldn't they have that option? Why should they have to pay for 300 mile range and the extra batteries.
I have an office in Chicago and one in Atlanta and I'm familiar with both. Even if you're commuting from Aurora to downtown Chicago, then you'd still only be doing ~80 miles round trip every day. But frankly, if you're commuting that far then you should either get a Volt or a really efficient hybrid...or get a new job because driving 90 minutes both ways in traffic...SUCKS!!!
Posted by: DaveD | 05 May 2015 at 01:44 PM
Whether one thinks that a 300 mile EV is necessary or not it is a commonly stated criticism of EVs that they do not have that range and it is always implied by the media that is one major drawbacks of EVs that have consumers rejecting them. I personally agree that 120-150 miles is more than enough since I would prefer an economical EV where I didn't pay for 100% more battery that I would rarely use and would have to haul around constantly for no reason. However, I think differently than the majority of consumers who really do listen to the TV pundits, the internet trolls and other fake experts. That said, if one were to assume that Nissan or GM or others can make a 150 mile EV with current technology batteries, advancements such as a Si-S battery which has double the capacity and uses lower cost materials would allow a 150 mile EV to halve the battery size (gaining some range by loss of weight) and a more than 50% decrease in the cost of the battery. Which would be significant. I thin the Li2S cathode success here is a big break through. I suspect that manufacturing it at industrial scale has some hurdles yet as I think Li2S is a little sensitive to water and O2, so it may be necessary to do at least some of the manufacturing in an inert environment. If the coated Li2S can be handled in a dry environment then the hurdle is not too big, but if even the coated particle needs inert atmosphere then plants may need to be significant modified. Yet, it still may be worth the added capital expenditure given the possible gain in cost and energy density.
Posted by: Brotherkenny4 | 08 May 2015 at 06:59 AM
It may be more practical to have EV batteries manufactured in stackable modular units each with 100 mile range capacity. This is ideal for most commuters and the most efficient weight to energy density ratio. The ability to add another battery module for extended range journeys would remove the need for frequent charging stops. This would be possible with the higher energy density of LiS Graphene batteries and XG Sciences appear to be a step ahead towards volume production. See http://www.prweb.com/releases/2015/03/prweb12575091.htm
Posted by: Marcus | 08 May 2015 at 03:33 PM