Sodium Ion developments thread

[neptronix]

So if you haven't noticed, sodium ion batteries are rapidly advancing.
I'm beginning to think sodium ion batteries are the next big thing for ebikes, due to their lower cost and higher safety.

So let's kick off a thread about these, eh?

 

[JackFlorey]

So if you haven't noticed, sodium ion batteries are rapidly advancing.

Three years ago, Bluetti announced their NA300 sodium battery power bank. I kept checking in with them at trade shows.

A year ago I managed to find the R+D guy at a trade show and he told me "I've been cycling those things at home for a while now and they have problems. So we won't be releasing it any time soon."

I keep hoping they solve their problems and release it.

However for ebikes I don't see sodium having too much advantage over LFP other than operating/charging temperature range.

 

[neptronix]

Two EV models powered by sodium-ion batteries roll off line in China - Batteries News

Farasis Energy’s sodium-ion batteries currently in production have energy densities in the range of 140-160 Wh/kg, and the battery cells have passed tests including pin-prick, overcharging, and extrusion, according to the company.

Farasis Energy will launch the second generation of sodium-ion batteries in 2024 with an energy density of 160-180 Wh/kg, it said.

By 2026, the next generation of sodium-ion battery products will have an energy density of 180-200 Wh/kg.

200 wh/kg 2 years from now? not bad!

 

[neptronix]

Three years ago, Bluetti announced their NA300 sodium battery power bank. I kept checking in with them at trade shows.

A year ago I managed to find the R+D guy at a trade show and he told me "I've been cycling those things at home for a while now and they have problems. So we won't be releasing it any time soon."

I heard about that. That was in 2021?

However for ebikes I don't see sodium having too much advantage over LFP other than operating/charging temperature range.

I think they'll get interesting for ebikes once we get into the 200whrs/kg range and beyond.

 

[Hillhater]

I think they'll get interesting for ebikes once we get into the 200whrs/kg range and beyond.

And in the meantime Lithium will be improved such that the weight of a Sodium pack will remain double that of lithium !
That is a big factor for ebikes.
then there is the unknown price for these new developments ?….they will have to be very much cheaper than lithium to be attractive.

 

[neptronix]

And in the meantime Lithium will be improved such that the weight of a Sodium pack will remain double that of lithium !
That is a big factor for ebikes.

Maybe. Lately, sodium ion energy density is increasing at a rate much times faster than lithium has in the past. Outside of next generation lithium metal / solid state batteries, lithium battery density has been increasing at a crawl over the last decade. If this trajectory continues, sodium ion could be on par, density wise, with lithium sooner than both of us expect.

Of course solid state technology will be a curveball, but the current main problem with it is the cost to manufacture the cells. These cells will definitely be the whr/kg champion someday, but their economic disadvantage will slow their development.

then there is the unknown price for these new developments ?….they will have to be very much cheaper than lithium to be attractive.

There's many projections for the prices based on materials costs and manufacturing processes involved in making sodium batteries. Generally they put the near-future price of sodium ion as being 40% cheaper than lithium cells once manufacturing hits a roughly equal level of scale.

This is such a big improvement in cost that it could put electric cars very close to price parity to gasoline cars. If these costs are correct, we can see the economies of scale in manufacturing improve way faster than it did for lithium, which would lead to even more rapid improvements in cost and density.

If not for cars, sodium ion will probably be a hit in grid storage, which could have the same effect on boosting the technology's growth as mass adoption in cars.

 

[harrisonpatm]

If not for cars, sodium ion will probably be a hit in grid storage, which could have the same effect on boosting the technology's growth as mass adoption in cars.

As well as for DIY home storage. While I can see them having their place in cars as the energy density increases (there are plenty of people like me who value battery pack lifetime longevity over range on a single charge), I'm not sure they're ever going to be a big hit with ebikes and other lightweight vehicles. Very much looking forward to them showing up in the secondhand market and grabbing them for my house.

 

[Pajda]

Recently ordered samples of SIB cells in 18650 format from hakadibattery.com (it was the first link to 18650 size after 5s googling, So much for choosing this particular brand ) Both HE and HP designs. Let's see what they can do...

 

[jonescg]

I think Na-ion is a winner from a recycling perspective. The hydrometallurgy invariably uses NaOH to neutralise any acidic solutions, and that only makes isolating the lithium harder. So if you get rid of that constraint, the process is a doddle.

 

[neptronix]

BYD breaks ground on its first sodium-ion EV battery plant

The world’s largest EV maker, BYD, broke ground on its first sodium-ion battery plant this week. BYD is investing $1.4 billion (RMB 10 billion) with 30 GWh planned annual capacity.

You likely heard that BYD just topped Tesla in overall EV volume to become the largest electric car maker globally. However, BYD is also a top global battery manufacturer.

 

[Pajda]

A few thoughts on the topic of Sodium-ion batteries (SIB).

The first is from global/market perspective. It is good to acknowledge that a large number of us have a distorted view due to the fact that we consider vehicle applications as "entertaining", and despite this it is then difficult to see the broader perspective on the issue of energy storage use. If our goal is sustainability and slowing global warming, then the main focus should be on decarbonising industry and energy production. Unfortunatelly the main problem with that is that this issue is considered as "boring" in terms of the functioning of the ordinary human mind. Thus, I see the greatest prospect for SIB in the application of stationary battery systems, where this market will surpass mobile/vehicle applications by an order of magnitude in the future.

The second is from technical/practical perspective of mobile/vehicle apps. It is not easy to determine what is the optimal endurance or range value for mobile/vehicle applications. But we know quite precisely how much is the minimum value for practical use. If we choose the aforementioned area of road vehicles (cars), then here it is a value of 200 miles / 320 km according to the EPA or WLTP driving cycle (maybe some of you remember that for example Elon Musk or GM talking about this value just around 2010). If you want to achieve such a range in a standard passenger car, you need to achieve a volumetric energy density on a cell level of at least 350 Wh/l in prismatic or 450 Wh/l in cylindrical format (gravimetric energy density plays only a minimal role in this particular application). SIB technology today is still below 300 Wh/l in prismatic format and so it is still not practically usable. The same story was with LFP chemistry. LFP technology entered the mass market around 2010 with marketing shitstorm about miraculous technology for EVs, but the first practical vehicle application was demonstrated by Tesla just after 2020. The technical milestone was the recent advance in volumetric energy density of LFP made by Chinese manufacturers CATL and BYD, which have finally surpassed the above mentioned 350 Wh/l level and now are reaching 450 Wh/l in prismatic format.

 

[Hillhater]

LFP technology entered the mass market around 2010 with marketing shitstorm about miraculous technology for EVs, but the first practical vehicle application was demonstrated by Tesla just after 2020.

you are forgetting that BYD had LFP battery powerd EVs (e6 model) in mass production by 2009.

BYD e6 - Wikipedia

en.wikipedia.org

Also the 2011 Fisker Karma which used the A123 LFP cells.

(gravimetric energy density plays only a minimal role in this particular application).

that a very odd approach for a MOBILE applications , where power to weight ratios are a major criteria ?

 

[Pajda]

that a very odd approach for a MOBILE applications , where power to weight ratios are a major criteria ?

The point is that there is a very widespread phenomenon in technology, which I personally call the "obsession with the mass", which sums up the approach that by reducing weight, every problem can be solved.

A typical example where this approach clashes are ground wheeled vehicles (four-wheeled vehicles to be precise). Even if we have a battery with infinite gravimetric energy density (zero battery mass) we still have the mass of the vehicle and payload itself to set in motion. So for a standard passenger car you still need a minimum of ca 40 kWh of stored electric energy for a practical 200 mile / 320 km EPA range without any battery mass (ca 50 kWh with available battery technology). So the crucial problem in this particular app is where to install this amout of energy. We know that for standard passenger car the available volume for energy storage is limited from ca 200 l in A segment up to 400 l in luxury or big SUV segment. We also know the volumetric assembly efficiency of modern battery pack (today it is around 50 % with prismatic and 35 % with cylindrical cell format). It is therefore easy to calculate what cell-level volumetric energy density you need for the required vehicle range.

Completely opposite problem is an eVTOL. Here the crucial parameter is the gravimetric energy density (mass) when volumetric energy density almost does not matter.

 

[BlueSwordM]

Actually, modern leading edge battery packs are approaching 83% packing efficiency

https://carnewschina.com/2023/12/14/zeekr-launches-800v-golden-battery-can-charge-500-km-in-15-minutes-and-withstand-1000-c/

 

[Pajda]

Actually, modern leading edge battery packs are approaching 83% packing efficiency

https://carnewschina.com/2023/12/14/zeekr-launches-800v-golden-battery-can-charge-500-km-in-15-minutes-and-withstand-1000-c/

Actually, they don`t. Did I mention I don't like marketing? I have already been asked about this Zeekr parameter and I am pretty sure they are using marketing tricks. They simply took the term CTP literally and so only included the cells and their mechanical and electrical connections in this value. If you just look at a common battery module, for example from VW's MEB platform, this module itself has about 82% volumetric assembly efficiency, so there's no chance they included the front compartment with the junction box in that value. The real value of their VAE will be around 60%, which is still a great result by the way. The bigger 82kWh pack from VW MEB platform have 48% VAE.

And sorry for the offtopic.

 

[Pajda]

you are forgetting that BYD had LFP battery powerd EVs (e6 model) in mass production by 2009.

Yes I forgot about the BYD e6, but it's a very good example to confirm my thoughts. Have you ever seen its battery pack from 2009? I only found this video where you can clearly see how huge its battery pack was! Its volume is definitely well over 500l and there is no chance to fit this into any reasonable sedan or limousine. I found on aliexpress, that its LFP battery modules have volumetric density of ca 160 Wh/l.

 

[Hillhater]

We know that for standard passenger car the available volume for energy storage is limited from ca 200 l in A segment up to 400 l in luxury or big SUV segment

.. that seems to only be defined by the existing designs of passenger vehicles.
Why is it not possible to concieve a design where the space for a battery is a primary factor rather than any preconcieved vehicle asthetics ?
Further , total weight is still a major consideration for many reasons such as motor power, transmission strength, chassis strength, suspension design,braking, tyres, and even road construction .

 

[Pajda]

SIB samples in 18650 format from Hakadi battery (Shenzhen Zhonghuajia Technology Co., Ltd.) arrived yesterday. Started with initial tests. So far I can say that they have announced capacity and energy and arrived in an adorable white cardboard box with pictures of green leaves.

But there is a crucial question type: "To be, or not to be" for the members of the ES Forum...

What the heck is the proper DoD test range for SIB?

From Hakadi datasheets even for the 100% DoD are available two ranges 4.10-1.50V for HE model, and 4.00-1.80V for HP model. So I'll start the cycle life tests with these settings, but tell me what DoD voltage range use for "daily commute" e.g., the equivalent of "90%" DoD cycle life for LIB.

 

[neptronix]

Excellent information, thanks!
128whrs/kg is pretty nice for a first gen cell... this is better than a lot of lifepo4.

What a weird discharge curve these batteries have. I'd personally cut the discharge below 2v.

 

[harrisonpatm]

Excellent information, thanks!
128whrs/kg is pretty nice for a first gen cell... this is better than a lot of lifepo4.

What a weird discharge curve these batteries have. I'd personally cut the discharge below 2v.

View attachment 346468

From 4.1 to 2.6, it looks like these could be drop-in replacements for lead acid. Maybe at 4s? That'd certainly be a good use for some first-gen cells.

 

[neptronix]

Starter battery scenario:
4.1v max x 4 = 16.4v
2.0v max x 4 = 8v
A typical car alternator may produce 13.8v / 4v = 3.45v per cell which is 50% SOC

So for starter battery duty this would only be using half the SOC range. Due to the better temperature resistance you may need less battery in a starter battery configuration.

For a storage battery, you may actually want 3s..
3s would be 12.3v fully charged.. regular lead acid is fully charged at 12.3-12.6v or so.
So for 3s you would be advised to use a special charger that doesn't exceed 12.3v at any time.

Not a great overlap with lead acid voltages but good enough for some cases.

 

[Hillhater]

Oo, err !
I have never been happy with inr type cell discharge voltage ranges , 4.2 -3.0 v…
..but a 4.2-1.5v range is very problematic .
even if you only use 4.2-2.0v range , that would still halve the voltage from a pack at 60% discharge !
Effectively, that halves the power output available, and looses 40% of pack capacity !
Bring back those old RC lipo’s !

 

[JackFlorey]

What a weird discharge curve these batteries have.

Yeah, it's not great for power electronics. A flat discharge curve means you need semiconductors at a specific voltage always carrying (roughly) the same current for a given power. This curve means you have to accommodate a 3:1 voltage AND current range.

 

[neptronix]

It's an almost worst case voltage curve for an electric vehicle because the output voltage is eventually half of when fully charged.
But are all sodium batteries gonna be like this?

We'll find out on the next episode!

 

[harrisonpatm]

It's gonna be pretty baller for stationary DIY solar storage, maybe even grid level storage.