New PNNL electrolyte may enable use of lithium anodes in very high capacity advanced batteries
24 February 2015
Researchers at Pacific Northwest National Laboratory (PNNL) have developed a new electrolyte that allows lithium-sulfur, lithium-metal and lithium-air batteries to operate at 99% efficiency, while having a high current density and without growing dendrites that short-circuit rechargeable batteries. An open-access paper on their work is published in the journal Nature Communications.
“This new discovery could kick-start the development of powerful and practical next-generation rechargeable batteries such as lithium-sulfur, lithium-air and lithium-metal batteries,” said PNNL physicist Ji-Guang Zhang, corresponding author of the paper.
Lithium (Li) metal is an ideal anode material for rechargeable Li batteries due to its extremely high theoretical specific capacity (3,860 mAh g−1), low density (0.534 g cm−3) and the lowest negative electrochemical potential (−3.040 vs standard hydrogen electrode). Extensive attempts have been made to use Li as an anode in rechargeable Li batteries since the 1970s, but several seemingly insurmountable barriers, including dendriticLi growth and limited Columbic efficiency (CE) during repeatedLi deposition/stripping processes, have prevented their large-scale applications.
… Electrolyte is one of the most critical elements that affects the cycling stability of Li metal anodes. Aurbach et al. indicated that Li is thermodynamically unstable with any kinds of organic solvents. The interactions between electrolyte components and Li metal results in significant side reactions that not only lead to a low CE but also consume Li metal and the electrolyte. This produces a solid-electrolyte interphase (SEI) film that may eventually grow into a thick layer, leading to high-impedance-failure of the battery instead of a short circuiting failure due to dendriticLi growth. This phenomena becomes serious especially at high current densities.
… Here we demonstrate that the use of highly concentrated electrolytes composed of ether solvents and the salt lithium bis(fluorosulfonyl)imide (LiFSI orLiN(SO2F)2) results in the dendrite-free plating of Li metal at high rates and with high CE. This exceptional performance cannot be achieved when lower concentration electrolytes are used (with or without LiFSI) and when LiFSI is substituted with other salts.
—Qian et al.
The researchers explored the morphology of Li deposition in different electrolytes using coin-type Cu|Li cells. After the initial deposition, the coin cells were disassembled to collect the Li films deposited on the Cu substrates for microscopic analysis by scanning electron microscopy (SEM) without exposing to air. In cells with the ether-based LiFSI electrolytes, the anode developed a thin, relatively smooth layer of lithium nodules that didn’t short-circuit the battery instead of growing dendrites.
After 1,000 repeated charge and discharge cycles, cells with the new electrolyte retained 98.4% of the initial energy while carrying 4 mA cm-2. The researchers found that greater current densities resulted in slightly lower efficiencies. For example, a current density as high as 10 mA cm-2, the test cell maintained an efficiency of more than 97%. And a test cell carrying just 0.2 mA cm-2 achieved 99.1% efficiency. Most batteries with lithium anodes operate at a current density of 1 mA cm-2 or less and fail after less than 300 cycles.
Results indicated that 4 M LiFSI inDME is an optimized salt concentration to obtain the stable cycling of Li metal in this electrolyte system.
The reactivity of these electrolytes is low (resulting in very limited side reactions and thus a high CE) and the large amount of Li+ cations available enables high current densities to be used forLi metal deposition. For more dilute electrolytes, the solvent is found to react with the plated Li metal to a much greater extent, which lowers the CE of the Li plating/stripping. Although the thickness of the SEI layer still grows with increasing cycle numbers as a result of the non-perfect CE (~99%) for Li plating/stripping in the 4M LiFSI-DME electrolyte, the highly conductive nature of the SEI layer leads to a highly stable voltage profiles during the cycling of Li electrode. In addition, the highly compact feature of the SEI layer also prevents further corrosion of the Li metal electrode and results in excellent stability of the electrode in the highly concentrated LiFSI-DME electrolytes.
—Qian et al.
The electrolyte needs to be refined before it’s ready for mainstream use. Zhang and his colleagues are evaluating various additives to further enhance their electrolyte so a lithium battery using it could achieve more than 99.9% efficiency, a level that would be needed for commercial adoption. They are also examining which cathode materials would work best in combination with their new electrolyte.
The new electrolyte’s remarkably high efficiency could also open the door for an anode-free battery, Zhang noted. The negative electrodes in today’s batteries consist of metal current collectors (e.g., copper) coated in active materials such as graphite or lithium. An electrolyte with more than 99% efficiency means there’s potential to create a battery that only has a negative current collector, without an active material coating, on the anode side.
Resources
Jiangfeng Qian, Wesley A. Henderson, Wu Xu, Priyanka Bhattacharya, Mark Engelhard, Oleg Borodin & Ji-Guang Zhang (2015) “High Rate and Stable Cycling of Lithium Metal Anode,” Nature Communications doi: 10.1038/ncomms7362
Hurry-up, my next car purchase is in 10 years and I want to have many great choice to choose from at that time. If this battery is good and gas remain low and they also increase mpg in gasoline car and they invent synthetic low cost gasoline, than I will have many great choice to choose from in 10 years. I want that all these products come on the market now, so in 10 years I will buy used for lower price possible while maximizing quality and efficiency.
Posted by: gorr | 24 February 2015 at 12:03 PM
97% initial capacity retention after 1,000 cycles is more than adequate. Batteries that will never cycle out are ideal for grid storage, lowering the cost per kWh stored when amortized over 10+ years.
Posted by: Anthony F | 24 February 2015 at 12:30 PM
Let's fine tuned and mass produce it by 2020 or before for 500 miles extended range BEVs?
Tesla/Panasonic should buy the rights and mass produce this high performance battery in their mega factory ASAP.
Posted by: HarveyD | 24 February 2015 at 02:33 PM
This is perhaps the most exciting BEV research announcement since Silicon Nanowires...except even more so.
1. It seems to solve the dendrite problem that high-theoretical-density chemistries such as Li-Sulfur and Li-Air have.
2. It seems to allow applicable chemistries to charge faster and harder.
3. It seems to offer a more compact and cheaper packaging.
4. It is offered by a reputable national laboratory.
We need tests of large-form-factor batteries suitable for BEV ASAFP.
Posted by: HealthyBreeze | 24 February 2015 at 04:47 PM
Agreed, if this solves the remaining problems of Li-S, let alone Li-Air, then this is the biggest news GCC has reported over the last year. If it is eventually commercialised, doubtless the first commercial format will be 18650, which means we can expect Tesla to be first to test it for EV use.
Posted by: clett | 25 February 2015 at 07:34 AM
Another group at Georgia Tech had already published a similar finding on December 10, 2014:
In Situ Formation of Protective Coatings on Sulfur Cathodes in Lithium Batteries with LiFSI-Based Organic Electrolytes. Adv. Energy Mater. 2014.
DOI: 10.1002/aenm.201401792
Gleb Yushin, Georgia Institute of Technology, Atlanta, USA, and colleagues used a solution of lithium bis(fluorosulfonyl)imide salt (LiFSI) in dimethoxyethan (DME) as an electrolyte for Li-S batteries. This electrolyte leads to a cathode mass loss of only 23 % after 1000 cycles and a high average coulombic effectiveness of nearly 100 %.
Posted by: Tray Mark | 25 February 2015 at 08:21 AM
There is SO much research going on that the average annual improvement in energy density is going to go from 7-8% historical numbers to more like 15% over the next 10 years.
There was just never the incentive to put anything into big leaps until the last few years and it takes a while to get them to market. So now, the pipeline of real advances are finally starting to come to fruition.
I can't wait to see what we have in 2020.
Posted by: DaveD | 25 February 2015 at 10:20 AM
For years, we have read about these 10X magnitude gains for anode, cathode, and now the electrolyte.
If true, pour in this electrolyte instead of present electrolyte and put the darn battery in public hands.
@Tray Mark notes this as a cathode coating was discovered at Georgia Tech last year.
Posted by: kelly | 02 March 2015 at 07:33 AM
Due to patent rights, it is almost impossible to use the best known elements in the same battery.
The world is domed to use compromises and we may never get the best battery to the market place, unless somebody is ready to challenge patent rights.
Posted by: HarveyD | 03 March 2015 at 07:40 AM