Why do lithium batteries die and how to improve them?

palmermd said:

I agree that they could not change the chemistry at that time, but you can work to optimize what you have.

Click to expand...

Given what we know about the LEAF batteries now, it appears to me that even thermal cooling would not have been enough to make the batteries last much longer than they do now for places like Los Angeles, unless they would be willing to sacrifice a large amount of efficiency (and range) to do so.

Look at the Volt TMS which uses a very similar battery as the LEAF and what maximum temperatures it aims for - it generally keeps the pack between 72-90F when the car is either on or charging. Even in Arizona, the Volt's battery appears to be holding up pretty well (hard to tell how much GM allows the SOC band to expand to compensate for capacity loss - don't know if people are able to monitor Volt battery pack voltages to get an idea or if they are able to sniff the CAN bus to find the data).

But LEAFs subjected to the same temperature band (like my San Diego LEAF which very rarely sees any battery temperatures over 85F) has still lost capacity at an alarming rate and TMS would not have helped my battery significantly.

It seems that GM got a battery from LG which has some "special sauce" which helps it's LiMn batteries last longer than the LEAF. And hopefully Nissan has their own "special sauce" ready to go soon.

drees said:

palmermd said:

I agree that they could not change the chemistry at that time, but you can work to optimize what you have.

Click to expand...

Given what we know about the LEAF batteries now, it appears to me that even thermal cooling would not have been enough to make the batteries last much longer than they do now for places like Los Angeles, unless they would be willing to sacrifice a large amount of efficiency (and range) to do so.

Look at the Volt TMS which uses a very similar battery as the LEAF and what maximum temperatures it aims for - it generally keeps the pack between 72-90F when the car is either on or charging. Even in Arizona, the Volt's battery appears to be holding up pretty well (hard to tell how much GM allows the SOC band to expand to compensate for capacity loss - don't know if people are able to monitor Volt battery pack voltages to get an idea or if they are able to sniff the CAN bus to find the data).

But LEAFs subjected to the same temperature band (like my San Diego LEAF which very rarely sees any battery temperatures over 85F) has still lost capacity at an alarming rate and TMS would not have helped my battery significantly.

It seems that GM got a battery from LG which has some "special sauce" which helps it's LiMn batteries last longer than the LEAF. And hopefully Nissan has their own "special sauce" ready to go soon.

Click to expand...

Dave, please don't forget that it's time spent at temperature, and not just the band. I'm sure that you know this. If we shaved off the peaks over 70 F in San Diego, particularly during charging, I would think that there would be an improvement in pack longevity. Additionally, Volt's pack is thermally insulated, and ambient heat does not seep in as rapidly as with the LEAF. Insulation in turn necessitates active cooling to remove the heat energy creted during operation of the vehicle. These two factors could be significant, and combined with a more robust cell design, and potentially an expanding SOC range, could provide a good end-user experience.

surfingslovak said:

Dave, please don't forget that it's time spent at temperature, and not just the band. I'm sure that you know this. If we shaved off the peaks over 70 F in San Diego, particularly during charging.

Click to expand...

Oh, I'm not forgetting. 98% of my charging is in the middle of the night when the pack has already cooled down to 75F during the summer. The pack typically hits a peak of 80F at the end of the day during the summer months - not quite hot enough to trigger cooling. It takes a heat wave to get temps higher than that. It's just not that hot here.

Yeah, if TMS activated during driving and cooled the pack down to 75F while driving that might have a nominal effect on battery life. Maybe I'd only be down 14% instead of 18%. But it wouldn't be anywhere near the improvement Nissan needs to make where they should be aiming for 20% loss after 5 years. For Arizona owners? Yeah, they might get close to So Cal durability.

In the winter months the pack rarely gets above 70F, so TMS would almost never be activated then.

drees said:

The pack typically hits a peak of 80F at the end of the day during the summer months - not quite hot enough to trigger cooling. It takes a heat wave to get temps higher than that.

Click to expand...

Please note that this is at the temperature sensors, which are attached to the ends of a few modules. The Volt flows coolant along the faces of each module, thereby keeping all of the modules in a condition where they are all nearly isothermal. In the LEAF, there is certainly a temperature rise from where the temperatures are sensed to where the heat is dissipated. The magnitude of that temperature rise is something only Nissan knows, but it I suspect it is non-trivial. The article on separators that Guy linked recently said that large-format cells often have a 10C temperature gradient across the cell under normal operating conditions. Perhaps the rise in the LEAF is more than that.

Does it make a big difference? Perhaps. Developing cells which have a long life even in high temperatures is what is needed for the EV industry long-term.

I think the ev industry needs to follow dysons lead. When the portable vac has below optimal power out just shuts down rather then giving a suboptimal experience. Likewise, the battery should have a buffer to the upside which initially allows a little less then full charge, but you'd that as the battery ages. Then one the battery goes to low, the maintenance light goes on, and signals it's time for a repair (under warranty for 3 years) the service shop can then restore our replace the battery. The user experience on range then status consistent. There is a cost, but overall satisfaction should justify that cost via higher sales and incremental sticker price.

DougWantsALeaf said:

I think the ev industry needs to follow dysons lead. When the portable vac has below optimal power out just shuts down rather then giving a suboptimal experience. Likewise, the battery should have a buffer to the upside which initially allows a little less then full charge, but you'd that as the battery ages. Then one the battery goes to low, the maintenance light goes on, and signals it's time for a repair (under warranty for 3 years) the service shop can then restore our replace the battery. The user experience on range then status consistent. There is a cost, but overall satisfaction should justify that cost via higher sales and incremental sticker price.

Click to expand...

Congratulations: you just tripled the cost of a Leaf and now I can't afford it!
I don't think people want their car "shutting down" on the highway @60mph :shock:

Stanton said:

I don't think people want their car "shutting down" on the highway @60mph :shock:

Click to expand...

No, you definitely don't want that. I've been through this with another EV, where the engineers saw it fit to shut down the car when a battery sensor malfunctioned. I can confirm from my own experience that this is a horrible idea, and the other drivers who have been through this, concur!

DougWantsALeaf said:

I think the ev industry needs to follow dysons lead. When the portable vac has below optimal power out just shuts down rather then giving a suboptimal experience. Likewise, the battery should have a buffer to the upside which initially allows a little less then full charge, but you'd that as the battery ages. Then one the battery goes to low, the maintenance light goes on, and signals it's time for a repair (under warranty for 3 years) the service shop can then restore our replace the battery. The user experience on range then status consistent. There is a cost, but overall satisfaction should justify that cost via higher sales and incremental sticker price.

Click to expand...

No, it's simple. Just do these steps:

1. Give realistic expectations on how fast you might expect the battery to lose capacity based on use and climate. Provide an upper and lower bound that one might see based on various usage habits. Make sure the car tracks operating conditions accurately to help improve future expectations.
2. Give a max charge slider like Tesla and say how much using lower SOC will improve calender life.
3. User can slide the max charge slider up as capacity is lost to maintain a consistent range themselves.
4. Give a price for battery pack replacement once the pack drops below required capacity.

Really, all the magic is in #1. Provide realistic expectations and err on the side of under promising and over delivering.

Do that and people will be able to make their own educated decisions in terms of what is and is not acceptable.

drees said:

No, it's simple. Just do these steps:

1. Give realistic expectations on how fast you might expect the battery to lose capacity based on use and climate. Provide an upper and lower bound that one might see based on various usage habits. Make sure the car tracks operating conditions accurately to help improve future expectations.

Really, all the magic is in #1. Provide realistic expectations and err on the side of under promising and over delivering.

Do that and people will be able to make their own educated decisions in terms of what is and is not acceptable.

Click to expand...

+10,000
False marketing will accomplish nothing other than complete total destruction of your BRAND!!!!

drees said:

Really, all the magic is in #1. Provide realistic expectations and err on the side of under promising and over delivering.

Do that and people will be able to make their own educated decisions in terms of what is and is not acceptable.

Click to expand...

I like it. Keep things simple.

+10,001

TimLee said:

drees said:

No, it's simple. Just do these steps:

1. Give realistic expectations on how fast you might expect the battery to lose capacity based on use and climate. Provide an upper and lower bound that one might see based on various usage habits. Make sure the car tracks operating conditions accurately to help improve future expectations.

Really, all the magic is in #1. Provide realistic expectations and err on the side of under promising and over delivering.

Do that and people will be able to make their own educated decisions in terms of what is and is not acceptable.

Click to expand...

+10,000
False marketing will accomplish nothing other than complete total destruction of your BRAND!!!!

Click to expand...

Great news:

Tesla Motor's Co-founder and Chief Technology Officer JB Straubel signed a research agreement with Dalhousie University's Jeff Dahn, Li-ion battery researcher with the Faculty of Science and his group of students, postdoctoral researchers and technical staff.

Click to expand...

http://www.powerpulse.net/story.php?storyID=32461;s=061820151" onclick="window.open(this.href);return false;

See also TSLA stock price discussion thread <-- Link jumps to the appropriate post of June 17, 2015.

keydiver said:

Great news:

Tesla Motor's Co-founder and Chief Technology Officer JB Straubel signed a research agreement with Dalhousie University's Jeff Dahn, Li-ion battery researcher with the Faculty of Science and his group of students, postdoctoral researchers and technical staff.

Click to expand...

http://www.powerpulse.net/story.php?storyID=32461;s=061820151" onclick="window.open(this.href);return false;

Click to expand...

Can we say that from what we know now about the Leaf battery pack, Elon Musk was right?
Tesla Motors (s TSLA) CEO Elon Musk has derided Nissan’s battery pack, which uses an air cooling system, as “primitive” compared with the sophistication of even Tesla’s first prototype, which uses liquid cooling. As a result, the LEAF pack will have temperatures “all over the place,” claimed Musk, causing it to suffer “huge degradation”

keydiver said:

Great news:

Tesla Motor's Co-founder and Chief Technology Officer JB Straubel signed a research agreement with Dalhousie University's Jeff Dahn, Li-ion battery researcher with the Faculty of Science and his group of students, postdoctoral researchers and technical staff.

Click to expand...

http://www.powerpulse.net/story.php?storyID=32461;s=061820151" onclick="window.open(this.href);return false;

Click to expand...

It seems this arrangement will pay off rather quickly. According to this article, Dahn and his team have already published a paper which should help pave the way to operation of Li-ion batteries at higher voltages, thus increasing the energy density:

electrek said:

The paper states that ethylene carbonate (EC), an organic compound until now thought to be an essential electrolyte component for Li-ion cells, is “actually detrimental for cells at high voltages” – 4.5 V and above. Tesla/Panasonic’s current cells are believed to be around 4.2 V, but the upcoming cells to be produced at the Gigafactory are expected to have a higher voltage.

In order to replace EC in the electrolytes, the group developed “EC-free linear alkyl carbonate electrolytes” and tested them in pouch cells. They report “excellent charge-discharge cycling and storage properties” and better cyclability of cells than cells with EC.

The researchers are hopeful that it could lead to better high-voltage battery cells:

“Further optimizing these linear alkyl carbonate electrolytes with appropriate co-additives may represent a viable path to the successful commercial utilization of NMC/graphite Li-ion cells operated to 4.5 V and above.”

Click to expand...

This two-year-old article in Quartz gives a good overview of the issue with operation of NMC at high voltages that Dahn is trying to address:

Quartz said:

The key to the second-generation cathode is an unusual jolt of voltage that unlocks the higher performance. The best current lithium-ion batteries operate at an average of 3.7 volts or so. The idea with the NMC formulation was to operate it at 4.7 volts. The resulting leap in energy is what got the licensees so excited.

But researchers began to notice a problem with the cathode at the higher voltage, which was an unexplained fast and steep loss of energy as the battery went through charge-and-discharge cycles. They called the malady “voltage fade.” Until it was fixed, they said, the second-generation material could not be used commercially.

Click to expand...

If the average operating voltage of NMC Li-ion cells could be increased from 3.85V to something like 4.3V, that would result in a significant boost in capacity. In addition, it would allow the same operating voltage to be achieved with a smaller number of cells.

I have to give Dr. Dahn a lot of credit here: he has had faith in his invention of 16 years ago based on its promise and has worked to eliminate the problems with its commercialization. He is a significant contibutor to the BEV revolution which is currently ongoing.

RegGuheert said:

This two-year-old article in Quartz gives a good overview of the issue with operation of NMC at high voltages that Dahn is trying to address:

Quartz said:

The key to the second-generation cathode is an unusual jolt of voltage that unlocks the higher performance. The best current lithium-ion batteries operate at an average of 3.7 volts or so. The idea with the NMC formulation was to operate it at 4.7 volts. The resulting leap in energy is what got the licensees so excited.

But researchers began to notice a problem with the cathode at the higher voltage, which was an unexplained fast and steep loss of energy as the battery went through charge-and-discharge cycles. They called the malady “voltage fade.” Until it was fixed, they said, the second-generation material could not be used commercially.

Click to expand...

If the average operating voltage of NMC Li-ion cells could be increased from 3.85V to something like 4.3V, that would result in a significant boost in capacity. In addition, it would allow the same operating voltage to be achieved with a smaller number of cells.

I have to give Dr. Dahn a lot of credit here: he has had faith in his invention of 16 years ago based on its promise and has worked to eliminate the problems with its commercialization. He is a significant contibutor to the BEV revolution which is currently ongoing.

Click to expand...

voltage fade and electrolyte degradation are 2 separate issues. A high voltage spinel like Ni.5Mn1.5O4 has nil voltage fade, but suffers from electrolyte degradation (and resulting cathode leaching). Similarly, high voltage bland NMC has nil voltage fade but still has the same electrolyte degradation (but less cathode leaching). Mn rich, Li rich NMC has voltage fade + electrolyte degradation. That class of cathode probably exhibits voltage fade even at conventional voltages levels, its related to how its high capacity Mn transitions to a high stability Mn during use.

recently it seems the Chinese have also cracked the higher voltage EC free electrolyte challenge, and using only conventional ingredients. This is massive news as it ushers in both a new class of cathode (high voltage Mn Spinel, cheap and safe) as well as promotes higher capacity in conventional NCA and NMC.

Just for reference so that we can try to use similar terminology, here is a link to a transcript a blogger made to Professor Dahn's lecture on Li-ion battery degradation.

ydnas7 said:

voltage fade and electrolyte degradation are 2 separate issues.

Click to expand...

This is where I am getting a bit confused. In the lecture, Professor Dahn describes two mechanism that are degrading the batteries: one on the anode and one on the cathode. At the anode, he describes the formation of a film which he calls solid electrolyte interphase (SEI), which apparently is a reaction between the graphite and the electrolyte. That causes a slippage of the voltage curve to the RIGHT. At the cathode, he describes an oxidation of the electrolyte which occurs due to the high voltages present. This results in additional reactions which ALSO causes a shift in the voltage curves to the RIGHT.

So the question is: What are you referring to as "voltage fade" and what are you referring to as "electrolyte degradation"? I'll guess that electrolyte degradation is the oxidation of the electrolyte at high voltages that appear at the cathode. But the slippage of the voltages to the right can hardly be described as "voltage fade" since they are increases in voltages. Perhaps it is considered "fade" because most battery chargers (like the one in the LEAF) use a constant voltage to determine when to terminate the charge cycle and this shifting of the voltage curve to the right results in an increasing inability to fully charge the battery?

ydnas7 said:

A high voltage spinel like Ni.5Mn1.5O4 has nil voltage fade, but suffers from electrolyte degradation (and resulting cathode leaching). Similarly, high voltage bland NMC has nil voltage fade but still has the same electrolyte degradation (but less cathode leaching). Mn rich, Li rich NMC has voltage fade + electrolyte degradation. That class of cathode probably exhibits voltage fade even at conventional voltages levels, its related to how its high capacity Mn transitions to a high stability Mn during use.

Click to expand...

Am I correct in assuming that it is this last class of Mn-rich, Li-rich NMC that is being discussed in the recent Dahn paper and the Quartz article?

ydnas7 said:

recently it seems the Chinese have also cracked the higher voltage EC free electrolyte challenge, and using only conventional ingredients. This is massive news as it ushers in both a new class of cathode (high voltage Mn Spinel, cheap and safe) as well as promotes higher capacity in conventional NCA and NMC.

Click to expand...

Unfortunately, the state-of-the-art in the world of computer modelling of chemistry is NOWHERE near having the ability to model something as complex as the electrodes in a Li-ion battery, so it seems that the work is still proceeding emperically. The good news with Li-ion is that there are so many eyes on the problem now and advancements such as Professor Dahn's extremely-accurate Coulomb counter GREATLY speed up the time between experiment and result.

Here are a couple more articles which reference Dr. Dahn's new paper:

Green Plug District: Ethylene Carbonate-free Electrolytes Promise Less Oxidation and Better-performing Li-ion Cells

Green Car Congress: Dahn team develops ethylene-carbonate-free electrolytes for better-performing high-voltage Li-ion cells

The Green Car Congress article contains the following results of testing various additives to 4.5V from the paper:



Quote from the paper (from the article):

Xia et al. said:

The work in this paper suggests that EC itself is the root cause of many issues associated with the operation of NMC/graphite cells to high potential. Electrolyte oxidation reactions at high voltages cause gas evolution and impedance growth, leading to cell failure. These parasitic reactions become very problematic at 4.5 V even with state of the art electrolyte additives PES211 in EC:EMC electrolyte. … This work demonstrates that cyclic carbonates such as VC, FEC and DiFEC can act as the enablers for EMC-based electrolytes which function well in NMC442/graphite cells tested up to 4.4 or 4.5 V.

Click to expand...

Interestingly, the article indicates that this work was NOT funded by Tesla, but rather by 3M:

Green Car Congress said:

3M has filed a patent on this work.

Click to expand...

RegGuheert said:

...Interestingly, the article indicates that this work was NOT funded by Tesla, but rather by 3M...

Click to expand...

He's only been working for Tesla since June, so not much time to do enough work to publish a paper. His announced move was about 18 months or so before then, and he had a contract with 3M to complete before he moved over to Tesla.