Li-Metal Battery Breakthrough? 500 Wh/kg Goal Draws Near

JUL 19 2018 BY MARK KANE 46

The new non-flammable fluorinated electrolyte opens the way for 500 Wh/kg energy density and very long-range electric cars.

The researchers at the University of Maryland (UMD), the US Army Research Laboratory (ARL), and Argonne National Laboratory (ANL) described in a Nature Nanotechnology article a non-flammable fluorinated electrolyte that can handle Li-metal anodes and the most aggressive and high-voltage cathodes.

The Li-metal anodes and high-voltage cathodes offer high energy density, but there was no way to solve low efficiency and short cycle-life. Now, according to test results, there is a chance to achieve 93% of capacity retention after 1,000 cycles (Li-metal anode and LiCoPO4 cathode) or 90% of capacity retention after 450 cycles (Li-metal anode and NMC811 cathode).

The level of 500 Wh/kg using high capacity anodes and high voltage cells (up to 5 V) is something that could become the final blow to range anxiety (nearly doubles the state-of-the-art commercial cells), but as always, there are so many battery breakthroughs reported that we don’t know if we should keep our fingers crossed on this one, or toss it in the ain’t-gonna-happen bin.

“… Here, we report a non-flammable electrolyte that demonstrates excellent stability toward both a Li-metal anode and high-voltage/ high-capacity cathodes. It consists of 1M lithium hexafluorophosphate (LiPF6) in a mixture of fluoroethylene carbonate/3,3, 3-fluoroethylmethyl carbonate/1,1,2,2-tetrafluoroethyl-2′,2′, 2′-trifluoroethyl ether (FEC:FEMC:HFE, 2:6:2 by weight). Unlike the previously reported fluorinated electrolytes, which suffered from increasing impedance at the anode side, this all-fluorinated electrolyte enables a high Li plating/stripping Coulombic efficiency of 99.2% and suppresses dendrites without raising the interfacial impedance. It also supports the stable cycling of NMC811 (Coulombic efficiency of ~99.93%) and LCP (Coulombic efficiency of ~99.81%) cathodes by forming a highly fluorinated interphase with thickness of 5–10nm that is responsible for the effective inhibition of electrolyte oxidation and transition metal dissolution. Unprecedented cycling stabilities were obtained for both Li||NMC811 (90% retention at the 450th cycle) and Li||LCP cells (93% retention at the 1,000th cycle).”

Source: University of Maryland via Green Car Congress

Categories: Battery Tech

Tags: ,

Leave a Reply

46 Comments on "Li-Metal Battery Breakthrough? 500 Wh/kg Goal Draws Near"

newest oldest most voted

The rule of thumb for BEV batteries is they need to withstand a minimum of 2000 cycles (to 80% of original capacity); PHEVs need even more.

Sorry, but 1000 cycles still ain’t good enough for commercial EV batteries.

Did you say “ain’t Goodenough”?

Dr. Goodenough might find 1K cycles a Good Start, for EVentual commercial EV batteries and adoption. The UMD development cycle has probably NOT run its full course, for this “all-fluorinated electrolyte” chemistry.

“93% retention at the 1000th cycle” is decent, and not overly excessive degradation, when compared to a non-TMS Leaf battery, which after three years (approx. 1000 cyc.), is in roughly the same capacity loss % pattern.

Typical Leaf battery at 3 yr/ 36k mi., is roughly at about 10% capacity loss ( + or – 2-3%) depending on DC fast charges.

[quote]“93% retention at the 1000th cycle” is decent, and not overly excessive degradation…[/quote]

Yes, okay, my mistake for not reading the article carefully enough. The rule of thumb is 80% at 2000 cycles, so 93% at 1000 should be fine.

We don’t know how these will perform at 2000 cycles. They could still be better than 80%.

I find the blind focus on cycles puzzling. If one cycle is 1000 miles, 1000 cycles are a million miles. Where is the problem with having a range decrease from 1000 miles to 900 miles after beating a car one million miles?

1000 miles on a single cycle? Hmm.

This is one reason the new Tesla Roadster has 620 mile range, though. 200k miles is only 325 cycles, so it can use one of these lithium metal chemistries.

These are 1000 deep cycles.

With larger batteries becoming standard, people will hardly ever run down the battery to 0 only charge to 100% when going on a long trip. In practice, they will show much much less degradation.

1000 cycles are the standard. Teslas were rated at 500 cycles. 1000 full cycles means 200 000 miles on a 200 mile EV, but since they can do much more partial cycles it is more like 400 000 miles, more than anyone would ever need. And that is of course just down to 80%, it’s not like it just dies after that.

The average car doesn’t even reach 150k miles before being scrapped.

Yeah especially if the battery capacity gets much bigger. Assuming (say in 10 years), the typical BEV goes 500 miles on a full charge, after 500,000 miles the car would go 450 miles.

That’s good enough performance for traveling salesmen and maintenance people. Now if the cost comes down also (rather like what happened to hard disks in computers), that would far and away convince people who still had nagging doubts about electric vehicles.

It would also minimize the need for on-peak fast charging – something that to date has not been very grid friendly. Of course, then EVERY HOTEL would have to offer decent overnight charging facilities for all of their electric vehicle guests, say, 15 kw for 100 cars simultaneously would still be a pretty impressive undertaking, (like 2 super wallmarts per hotel).

It depend how they count a “cycle”.

What I understand about that metric, is it’s a full 100% cycle, something that every BMS (not TMS) is avoiding when controlling a somewhat expensive battery.
They charge it full and depleted until it’s dead, rince and repeat for 1 000 times and then measure what the battery hold the last one.

Since everybody get at destination with some energy left and many time leaving with a battery that has not completly charge, how much cycle does that count, 75%, 50% or much less.
LeafSpy track the number of recharge since late 2013 and on my 2012MY Leaf, last time I check, it has been done over 8 000 times on L1/L2 charging and it’s still going, althought not as far with a SOH of 76%.

No, the rule of thumb is (or at least use to be) an 80%/20% cycle when testing for longevity. Cycling li-ion batteries to 100% or 0% repeatedly wears them out very quickly. Typically batteries would wear out in a few dozen, or at most a few hundred, cycles to 100% and/or 0%.

That’s true only in the sense that a cell’s minimum and maximum voltages, as specified by the manufacturer, are not the physical limit, but rather some (somewhat arbitrary) values that result in the rated cycle life. However, for all practical purposes, these specified limits are what is considered 0% and 100% state of charge.

Uh, no.

1,000 cycles on a full sized EV battery (250 mi/charge) is 250,000 miles. And, according to the report, is only to around 90%. That is plenty. If they can do 500Wh/kg, and assuming a commensurate increase in Wh/l (volume), you’re likely to see packs that deliver closer to 300-350 miles of range instead of 250.

Now if we’re talking PHEV, you might need more cycles. My Volt is probably somewhere around 2,000 cycles (of the usable range, less of the full 16kwh capacity) and I feel like its starting to lose some range.

Hybrid vehicle batteries are cycled less deeply, to prevent too much degradation due to the more frequent charges/discharges.

So exciting the 25th announcement of a battery breakthrough this year!

We really need a 1.000 per year. Then if 0.1% make it to production, we could be getting somewhere 😉

Hmmm, yes. We’ve seen an acceleration from about one breathless, wide-eyed announcement about a battery breakthrough every other week, several years ago, to now about one per week.

Too bad the improvements in commercially produced li-ion batteries has not similarly accelerated.

Most of the “breakthroughs” promise some revolutionary new chemistries or approaches, that usually turn out to not be practicable due to other constraints, or still need a decade of further research to make them practicable.

This research however is addressing a very specific concern with well-known approaches. It’s precisely the kind of research real battery makers are doing all the time, to give us regular incremental capacity increases, or to bring new cell types to market.

Of course, this one research paper won’t suddenly give us batteries with twice the capacity overnight — but it shows a promising path towards overcoming the main issue holding back major capacity increases (and without any fancy solid-state electrolytes, too!), that could become available in commercial cells in a not too distant future.

Fluorinated….they ought to call the new electrolyte “Crest”. Recommended by 4 out of 5 battery engineers to reduce cell decay by 34%;)

Awesome. I will use this to fuel my car, which runs on battery breakthroughs. I get at least two refills a week.

You all complain when they don’t give you enough information and you complain when they give you plenty. I for one hope that Insideevs continues to report on battery “breakthroughs” even though they may be dead-end ones. Continued research is vital to the success of evs. Eventually, one of them may succeed, and it would be great if I read it first in Insideevs.


Indeed. Report on everey interesating battery technocology, and time will tell if any of them pans out. We don’t have to vote on the future battery here and now.

Reporting on *every* one could be pretty tedious, considering how many of them get published all the time… What bothers me though is that media tend to occasionally pick out some report seemingly at random, as if they were particularly groundbreaking, although in reality they usually aren’t any more promising than all the others that go unreported. It doesn’t provide a good picture of what venues are most relevant in ongoing research.

This one is a good pick though IMHO.

I can’t begin to guess exactly what the BBB (big battery breakthrough) will be, or exactly when it will arrive.

But given the financial incentives and the number of corporations and universities chasing BBBs, I’m highly confident we’ll see a series of advancements and a few blockbuster BBBs (BBBBs?) in the next 5 to 10 years.

So on one hand I’m quite confident we’ll get there. Some research team somewhere will find a battery formula that results in a combination of price and performance that, by today’s standards, sounds like bad science fiction.

But I’m at least as impatient for it to finally freaking happen as anyone here. I’ve been following this stuff since I read about what we now call a PHEV, built by some tinkerer in his garage, in Popular Science in the 1970s (1974, I think).

Don’t hold out for some big breakthrough. The real value of such reports is showing that there are enough venues left to explore, that we can realistically hope for the improvements that happen regularly to keep coming for quite some time 🙂

Each finding points in a direction when you know the reason for the result.

Ron Swanson's Mustache

IIRC, Elon claims that at 400 Wh/kg, generalized electric aviation becomes possible. If these guys are getting 500 Wh/kg, that is truly outstanding.

Of course, even if it works well and holds up to multiple charge/discharge cycles, that’s no guarantee that the cells can be manfuctured at a price point that is affordable.

Those guys look like they could star in an awesome buddy comedy.

This isn’t one of those vague breakthroughs we keep hearing. It’s an incremental improvement addressing a known issue with a well-studied battery design. Whether this advance gets commercialized or not, it’s a sign lithium metal anodes are maturing.

Solid Energy sells a 500 Wh/kg cell now. Cycle life is too low for EVs, but just wait. Cell phones and tablets in a year or so, high end EVs like the new Roadster a year after that.

93% retention over 1000 cycles is remarkable but not a breakthrough. Recent real use data already shows retention of over 90% at 150K miles, so existing battery technology is already there or near there. High voltage cell can also be a big factor in battery longevity.

I think the high capacity (500 Wh/kg or 100kWh pack at 200kg plus packaging) is the key to achieving cost and range parity with ICE. Once that happens, game over for ICE.

Not sure what the charge rate looks like based on the info provided.

Of course a cycle number in isolation is meaningless. The point is reaching such cycle stability with these high-capacity electrodes, which is indeed a breakthrough as far as I’m aware.

I think what mostly needs to happen is for the price to go down, and that has really been happening the last couple years, along with Wh/kg going up.

I understand that solid state batteries are basically inflammable and don’t require a thermal management system, which helps reduce costs even more. Solid state electrodes could be implemented for electric cars by 2020.

See this great article:

I expect they would need some heating, though?

Good question. I’ve tried to find more information online the last hour but haven’t been successful. I wish I knew more. Hopefully someone knowledgeable reads this.

Very likely will still need heating to keep the temperature above freezing, or at least not much below. Even polymer/solid-state li-ion batteries depend on chemical reactions to produce electricity; chemical reactions which are significantly slowed at lower temperatures. And the problem with plating when you try to charge frozen batteries may not be solved with solid-state tech, either.

Considering that lithium metal anodes rely on plating in their regular operation, I *think* this shouldn’t be a problem… Can’t cite any sources, though.

The discovery in this article is an example of why we may never see solid state batteries in mainstream EVs.

If we can get 500 Wh/kg with a non-flammable liquid electrolyte, what improvement will a solid electrolyte provide? They also have worse cold-temp performance, and aren’t as easy to process.

The rolled cells cranked out of the gigafactory have a lot of room for improvement as such innovations develop.

A more accurate count is how many mW passes through the battery. That way partial charges are counted as well. We sell stationary batteries and after looking at how everyone was counting charge cycles it was ridiculous. BMS can’t easily count charge cycles. what they can count is how much energy has passed through the battery.LG Chem measures their batteries the same way energy passing through the battery.

First of all, mW is a unit of power, not energy. You probably meant mWh.

More importantly, mWh is useless for comparing different cell types or chemistries, as it depends on the cell size. But once you know the size of the cell you are dealing with, you can do the conversion trivially.

500Wh/KG if that was at pack level, and not Cell level, it would mean my current Tesla 100kWh pack that weights # 700kg could be extended for the same weight to 350kWh capacity, that at just 1C could charge in theoretical # 1h on a 350kW/800V charger, and my car could run >1000 Miles with such a capacity !!! Has Wonderland arrived ?
Missing here is the volumetric capacity, is it also 500kWh/Liter or far less requiring a lot more space than per kg ?
Other missing is the “C” charge-level at which these life cycles were measured, to understand their ultra-fast chargin,g capability or if not possible. I’d like to know how many cycles they commit at real current Tesla 1.2C capped charging levels (80% charge in 45mn, 90% in 1h, and 99% in # 1.25h in real life), and at the up to # 3.8C expected for Porsche Tycan/Mission-E expected to charge 80% of # 92kWh in only 15mn at 350kW/800V although never demonstrated in real life video yet.
Why not giving all FULL SPECS in 1 GO ? Why hiding ? What is the always hidden issue here ?

What’s the point of comparing current pack level densities vs. future cell level densities?…

The cycle life was most likely at 1C, since that’s what’s most commonly used in data sheets.

In EVs designed from the ground up, gravimetric density (along with price) tends to be a more limiting factor than volumetric density. Also, the discrepancy between gravimetric and volumetric density tends to be negligible for cells of a similar type.

They didn’t give full specs, since this is a research paper — they are not talking about a specific cell model they want to bring to market. Things like C rating depends on the trade-offs taken in a specific cell model.

(Also, the full paper surely elaborates on the precise conditions they tested, like any reputable research institution would — they obviously can’t mention every single detail in the abstract…)

last 20 years of reading of “battery break through” . Never made it to real life……

Clearly you must have missed some of them, considering that several major advancements in Li-Ion batteries came to market over the past 20 years 🙂

(None of the modern chemistries, including NMC, NCA, LFP, or LTO, existed 20 years ago.)

Without funding even the best technology won’t make it to the market. Its not that because a breakthrough wasn’t commercialised that the trechnology is flawed.