Why do lithium batteries die and how to improve them?

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

DaveEV

Well-known member
Joined
Apr 23, 2010
Messages
6,253
Location
San Diego
Ran across this great presentation from Professor Jeff Dahn (Dalhousie University) talking about various issues in engineering batteries for EVs (and other applications) that last a long time. Professor Dahn has been working with GM for a year or two now helping them in their battery testing lab.

[youtube]http://www.youtube.com/watch?v=pxP0Cu00sZs[/youtube]

Notes:
  • Coulombic efficiency is an easy way to compare cycle/calendar life of a lithium batteries without having to run very prolonged cycling tests - the better the efficiency, the more durable the battery.
  • Chemistry (and additives) are vital in determining cycle/calendar life of a lithium battery.
  • The colder the battery, the longer it will last (Arrhenius' equation).
  • The lower the SOC the battery is charged, the longer it will last (example was storing battery at ~20% or 3.5V resting voltage).
  • LiMn (as used in the LEAF/Volt) has horrendous cycle/calendar life at temperatures of 30C+ compared to say LiCo (as used by Tesla). LiCo is even better than LiFe (A123).
  • Additives can have a very strong effect on coulombic efficiency (and thus durability)
 
Amazing...I just got done watching and came here to post a link and there it was already posted. Fantastic presentation.

Really shows why the guys in Arizona who charged more than once a day were having so much degradation. Also interesting his explanation of how Nissan missed it by doing the accelerated testing. Problems occur with the length of time you spend charging or discharging at high temperature. With the accelerated testing the cells did not spend much time at a high SOC. Actual use with a 3kW charger they spend a lot of time (relatively) charging. Discharge (driving) is probably similar to the accelerated testing rates.
 
Very impressive. 3 weeks to get predictive results instead of 8 years. The rate of improvement should.... improve.
 
All this begs the question, what did Nissan change in the chemistry of the "hot battery" to allegedly reduce degradation? And how successful will it really be in the real world?
 
drees said:
[*]The lower the SOC the battery is charged, the longer it will last (example was storing battery at ~20% or 3.5V resting voltage).

3.5V is pretty darn close to turtle. Is the implication that it's better to leave a Leaf sitting with 8 Gids vs 80 or 180?
We all know higher voltages are bad (although not nearly as bad as higher temperatures), but is there any accelerated degradation mechanism at 3.0-3.5V (turtle)?
 
GregH said:
drees said:
[*]The lower the SOC the battery is charged, the longer it will last (example was storing battery at ~20% or 3.5V resting voltage).

3.5V is pretty darn close to turtle. Is the implication that it's better to leave a Leaf sitting with 8 Gids vs 80 or 180?
We all know higher voltages are bad (although not nearly as bad as higher temperatures), but is there any accelerated degradation mechanism at 3.0-3.5V (turtle)?

Iirc, at very low SOC, the Copper anode starts to be attacked, enters the electrolyte and Copper plating starts ruining the cell. I don't know what the critical voltage that represents with the LEAF chemistry.
 
Nubo said:
Very impressive. 3 weeks to get predictive results instead of 8 years. The rate of improvement should.... improve.
Hopefully AESC is using this technique. At least we know that GM is.

TomT said:
All this begs the question, what did Nissan change in the chemistry of the "hot battery" to allegedly reduce degradation? And how successful will it really be in the real world?
So far all we know is that they are very likely going to a ceramic coated separator likely from Polypore. They did make slight chemistry tweaks for 2013, but who knows what other tweaks they have in store for the "hot battery". We need more data from our real world accelerated testing lab (also known as Phoenix).

GregH said:
3.5V is pretty darn close to turtle. Is the implication that it's better to leave a Leaf sitting with 8 Gids vs 80 or 180?
We all know higher voltages are bad (although not nearly as bad as higher temperatures), but is there any accelerated degradation mechanism at 3.0-3.5V (turtle)?
Yeah, the 3.5V probably shouldn't apply directly to the LEAF's batteries, as at a 3.5V resting voltage it's well below 20% SOC.

Somewhere around 3.75V resting voltage (~360V pack voltage) or around LBW would be a good suggestion for the LEAF, but realistically, anything between VLBW and 80% is good enough for short term storage. For longer term ideally you'd want somewhere between LBW and 40% and to keep it as cold as possible.
 
Nubo said:
GregH said:
drees said:
[*]The lower the SOC the battery is charged, the longer it will last (example was storing battery at ~20% or 3.5V resting voltage).

3.5V is pretty darn close to turtle. Is the implication that it's better to leave a Leaf sitting with 8 Gids vs 80 or 180?
We all know higher voltages are bad (although not nearly as bad as higher temperatures), but is there any accelerated degradation mechanism at 3.0-3.5V (turtle)?

Iirc, at very low SOC, the Copper anode starts to be attacked, enters the electrolyte and Copper plating starts ruining the cell. I don't know what the critical voltage that represents with the LEAF chemistry.
IIRC (and my recollection is foggy at best).. low voltage degradation is at extremely low voltages (0-2V) and not any voltage a Leaf battery would ever be allowed to see (3.0-3.5V.. although I have seen screenshots showing 2.7xV)
So I guess my question would be: Is it ANY worse (for battery longevity) to leave a Leaf sitting at 3.5V (say, 10-12 Gids or so) than at 50% SOC?
 
drees said:
Somewhere around 3.75V resting voltage (~360V pack voltage) or around LBW would be a good suggestion for the LEAF, but realistically, anything between VLBW and 80% is good enough for short term storage. For longer term ideally you'd want somewhere between LBW and 40% and to keep it as cold as possible.
But not VLBW (24 Gids)? Not 10-15 Gids?
If not, why not? (with respect to battery longevity)
 
drees said:
[*]LiMn (as used in the LEAF/Volt) has horrendous cycle/calendar life at temperatures of 30C+ compared to say LiCo (as used by Tesla). LiCo is even better than LiFe (A123).
[*]Additives can have a very strong effect on coulombic efficiency (and thus durability)[/list]
Thanks drees. Marvelous informative video. Everyone that wants to understand lithium ion batteries should watch this :!: :D

The truly horrendous coulombic efficiency of the LiMn battery as used by the LEAF and Volt does raise interesting questions about why Nissan and GM went that way.
LiMn does have advantages, but it is primarily in safety, markedly lower run away potential.
There haven't been any LEAF HV pack related fires, and only the one post accident Volt fire, where no one did the right thing and failed to remove the fire potential post accident.
Who would leave gasoline in a damaged vehicle :?: :?: :?:

While Tesla with 18650 commodity cells, which they took extensive efforts to protect and eliminate fire hazard, have still encountered a couple of unfortunate post road hazard / post accident fires.
And use of LiCo on an airplane (which in my own personal opinion was ill advised and which the FAA should still take a much harder look at) had serious problems.
Time will tell if Boeing has fixed the problem and adequately minimized the hazard.

GM did take the approach of providing temperature management for their LiMn version, which as the professor acknowledges in the video has resulted in much better capacity retention than the LEAF.
The real question is whether you can tweak the electrolyte chemistry with the right combination of trace additives to get a reasonable capacity life in a LiMn battery without a temperature management system :?:

Nissan may be correct that not having a temperature management system is the right design approach.
As Dr. Edward Buiel of Coulometrics pointed out in his presentation to the Chattanooga Engineers Club recently, a battery temperature management system can be a huge energy drain. Research on an electric bus a few years back showed the battery temperature management system taking as much energy as what the bus was using for propulsion :!:

But not having a temperature management system is only the right cost effective design approach if you have a battery that can maintain capacity for 100,000+ miles (and while being sufficiently safe). Nissan did NOT get the capacity side of the question correct for the product they have been selling in 2011 / 2012 / 2013 / early 2014. We can only hope they have come up with the magic five or more part additive for the electrolyte to get there with the HOT battery. And that they will offer it to buyers of the early defective HV pack LEAFs at a reasonable pro-rated / variable cost only price, and to the people they are now attempting to sell used LEAFs to that are coming off of lease. If they don't, they may have destroyed the Nissan EV brand :shock: :( :shock: :(
 
GregH said:
drees said:
[*]The lower the SOC the battery is charged, the longer it will last (example was storing battery at ~20% or 3.5V resting voltage).

IIRC (and my recollection is foggy at best).. low voltage degradation is at extremely low voltages (0-2V) and not any voltage a Leaf battery would ever be allowed to see (3.0-3.5V.. although I have seen screenshots showing 2.7xV)
So I guess my question would be: Is it ANY worse (for battery longevity) to leave a Leaf sitting at 3.5V (say, 10-12 Gids or so) than at 50% SOC?

The battery he mentioned sitting at 3.5v for 5 years was not necessarily a Leaf battery chemistry. In fact odds are it was not. His point was that he had a battery sitting at 20% SOC (which was 3.5v for that particular battery) and after all that time in a non temperature controlled situation the cell appeared to be "new" for all practical purposes.
 
Very Interesting presentation, even for a non-engineer like me. A lot of it was way over my knowledge and understanding, but I did get something from the charts. Very Interesting Indeed!

Now, could some of you who do understand all that was said please come up with some practical "do's and dont's" for the rest of us Leaf Drivers to minimize degredation of our battery pack, based on this study and video? It would be greatly appreciated.

Such as, only charge to 100% just before needing the extra charge, otherwise limit the charge to 80%.
If you need to store the Leaf for more that a couple of days leave the charge between 20% and 50%. Only use the 120v trickle EVSE if you do not have the availability to use 240v EVSE or DCQC, but minimize the use of the slow charging. Beware of any activity with battery temp above 30*c (86*f). In this regard, as I understand it, the Leaf uses air flow to cool the battery pack. If this is the case, could those in high temp environment utilize some sort of water misting system to cool the air moving through the battery pack? We would have to make sure there was water in our tank, but could it help reduce the temp?

These were a few things that I understood from the presentation. Please correct these if I have them wrong, and add to these if it would be helpful.

Thank you all for sharing your knowledge for the benefit of us all.
 
A lot of what you have suggested has already been tried.

Unfortunately, for the LEAF's battery, the degradation rate appears to be so overwhelmingly affected by the number of hours the battery temp spends above 25C (77F), that nothing else matters very much. Of course, increasing the temp to 30C or 35C just makes things worse. It was not unusual for my LEAF's battery temp to stay between 30C and 35C almost continuously from May through October.

I had the LEAF sitting idle in my driveway with a 50% charge for nearly the entire month of December and January and the battery continued to degrade. The simple reason was that air temperature was warm.
 
Graffi said:
VNow, could some of you who do understand all that was said please come up with some practical "do's and dont's" for the rest of us Leaf Drivers to minimize degredation of our battery pack, based on this study and video?

Temperature, Temperature, Temperature, especially while charging or discharging.

Park in as cool a place as you can. Use an end-timer to allow the car to charge when the battery is at its coolest. Avoid ChaDeMo when it's really hot. Drive gently when it's really hot. Avoid L1 charging.
 
Graffi said:
In this regard, as I understand it, the Leaf uses air flow to cool the battery pack. If this is the case, could those in high temp environment utilize some sort of water misting system to cool the air moving through the battery pack? We would have to make sure there was water in our tank, but could it help reduce the temp?
The only air flow for the LEAF battery is when the vehicle is moving. There are openings in the plastic cover that do cause some air flow past the under side of the space the battery box is sitting on while the vehicle is moving.
If outside temperature is less than the battery temperature, then there is some limited cooling.
If the outside temperature is higher than the battery temperature, the air flow is actually causing some limited heating of the battery.
In either case, I don't think water misting would have much impact on the heating or cooling unless you were cooling the liquid to below the outside ambient temperature, say using the LEAF air conditioning system.
Yes, people do use evaporative mist cooling fans in fairly dry areas, but most of the effect is because the water they are using is colder than the ambient air temperature.
 
Nubo said:
Avoid ChaDeMo when it's really hot.

Avoid charging at all when it is really hot. His measurements show that ChaDeMo charging when hot will likely be less detrimental than L2 charging because it take less time.

If I lived in Phoenix, I might just look into the "home 20k chademo" that some of the forum members are developing. Heck it might even be good to use around here in August.
 
palmermd said:
Nubo said:
Avoid ChaDeMo when it's really hot.

Avoid charging at all when it is really hot. His measurements show that ChaDeMo charging when hot will likely be less detrimental than L2 charging because it take less time.

Except ChaDeMo itself raises the LEAF battery temperature significantly if taken all the way. I guess I'd add a caveat about not trying to take DCQC to full SOC.
 
Nubo said:
Except ChaDeMo itself raises the LEAF battery temperature significantly if taken all the way. I guess I'd add a caveat about not trying to take DCQC to full SOC.

Yes. I only charge to 80% anyway. If the 20k home unit had a timer and could stop at 80% (or even better if it could stop any any preset the user selects) it would be great.
 
Awesome presentation. Thanks for finding and sharing it. On the positive note it's great to see that the tools to make faster progress are coming together.

On the negative side, the existence of the "other" failure mode, the catastrophic one was news to me. The degradation we're all talking about for the LEAF is the gradual one. What about the catastrophic one, does anyone know what should be our expectation about that?
 
Back
Top