SVOLT unveiled its new cobalt-free battery cell (update)
In an online presentation SVOLT gave us more information about its NMx cobalt-free battery cell that’ll arrive next year.
This long cobalt-free battery cell from SVOLT is specially made to be used in battery packs assembled with the CTP (cell-to-pack) technology.
With the CTP technology instead of having battery cells inside modules, then modules inside battery packs, we remove modules altogether. We end up with long prismatic battery cells connected in series that are put in an array and then inserted into a battery pack, making it as simple as it can be.
Let’s see some details about the battery cell.
Specs
- Capacity: 226 Ah
- Gravimetric energy density: 240 Wh/kg
- Volumetric energy density: 590 Wh/L
- Chemistry: LNMO
- Dimensions: 575 x 21,5 x 118 mm
Update
Initially when I wrote this article I thought that these SVOLT battery cells were made in high-voltage spinel form, but I was wrong. If true these battery cells would have a nominal voltage around 4,7 V and could be charged at 5 V. It would be possible to assemble simpler battery packs with fewer cells. Only 80 cells in series would be needed to reach 400 V, or 160 cells in series to reach 800 V.
Discharge profiles of different cathodes by BASF
However, after calculating the nominal voltage it’s clear that the LNMO battery cell from SVOLT isn’t made in a high-voltage spinel form. It has a nominal voltage of 3,8 V and a maximum charge limit of 4,3-4,35 V.
Anyway, regarding cost and energy density, LNMO battery cells offer a good balance when compared to alternative technologies.
NCM 811
- Gravimetric energy density: 270-300 Wh/kg
- Volumetric energy density: 620-700 Wh/L
- Cost per kWh: 90-80 euros
LNMO
- Gravimetric energy density: 240 Wh/kg
- Volumetric energy density: 590 Wh/L
- Cost per kWh: 80-70 euros
LFMP (high voltage version of LFP)
- Gravimetric energy density: 200-225 Wh/kg
- Volumetric energy density: 490-530 Wh/L
- Cost per kWh: 70-60 euros
Battery cell cost of different chemistries
What about longevity?
In the early days, some electric cars used manganese-rich battery cells that were extremely safe but weren’t very robust. At higher temperatures the manganese in the cathode tended to be corroded by the liquid electrolyte, which gradually reduced the ability to store lithium ions. We all remember the mess of the first generation battery packs in the Nissan LEAF made with LMO cells and without a proper TMS…
Fortunately this corrosion problem with manganese-rich battery cells was later solved by modifying the electrolyte and the coating of the cathode’s surface. This was the “secret” behind the more robust 2015 Nissan LEAF’s lizard battery or the GS Yuasa LEV50N cells used by Mitsubishi i-MiEV.
Modern manganese-rich battery cells are not only extremely safe, but also durable.
However, the arrival of more energy-dense NCM battery cells made us forget manganese-rich batteries, until now…
Manganese-rich batteries are coming back
SVOLT says that its new LNMO battery can deliver a 880 km range and lasts for 15 years and 1.200.000 km, which represents 1.500 cycles before reaching the EoL (End-of-Life). Some researchers consider that the EoL is reached when a battery only retains 70 % of the initial capacity, while others consider 80 %.
Given that these figures are likely in the fairy tale NEDC standard, we should get around 660 km of range and 900.000 km of service life with the more realistic WLTP.
Moreover, if we assume that the EoL happens at 70 %, it means that after 900.000 km the battery will still be able to deliver a WLTP range of 462 km.
Regarding cathode development of LIB (Lithium-ion batteries) we are currently at phase 2 with nickel-rich NCM 811 and upcoming NCMA battery cells, where costs are reduced by replacing as much cobalt as possible with nickel.
The phase 3 starts when more and more nickel is replaced with manganese.
To understand why phase 3 is important, let’s see the average market price of these raw materials per ton.
- Cobalt: 27.000 EUR/t
- Nickel: 11.000 EUR/t
- Manganese: 2.000 EUR/t
NCM Product Overview by BASF in November 4, 2014
Anyway, it’s great to soon have at least two compelling cobalt-free battery technologies available. It’ll be interesting to see which one will be favored by the Korean battery cell makers. Will it be LFMP or LNMO?
Finally, you can see the full video presentation of SVOLT’s cobalt-free battery and try to catch some details that I may have missed. The video has some interesting information but the content could be condensed in a 5 minutes video instead of 84.
https://www.youtube.com/watch?v=Amnu5FBL58g
More info:
https://blog.topsoe.com/the-cathode-material-for-next-generation-lithium-ion-batteries-is-ready
https://info.topsoe.com/batteries
https://www.mtixtl.com/EQ-Lib-LNMO.aspx
https://www.sciencedirect.com/science/article/pii/S2468025720300200
https://sci-hub.se/https://pubs.acs.org/doi/10.1021/acsami.6b02491
Charging voltage of 5V is amazing news for mobile gadgets. Usb-direct without transforming.
Good point, didn’t think of that.
That sounds interesting. Could you explain a bit more about that?
He can’t, because he doesn’t understand it, otherwise he wouldn’t wrote that nonsense
The charge voltage limit of current smartphones’ batteries is 4,2-4,3 V, but USB delivers 5 V, which means that the internal charging controller (buck converter) has to step down the voltage to keep the charging process safe.
However, with this new high voltage chemistry the charge voltage limit is 5 V, which means that it doesn’t need the buck converter active when charged by USB, making the charging process more efficient and safer (less heat generated).
Pedro this is utopia 🙂 You will never unify battery cells to use common 5V top charging voltage. The second problem is that USB “5V” is nominal value with huge tolerance (4.4-5.25V). So you still always need at least a voltage stabilizer. Also USB “5V” standard is limited to 2A. Majority of modern devices with USB fast charging (>10W) are using higher voltages like 9V or more (up to 20V) under QC(quick-charge) or PD(power delivery) or another standard.
Yes Pajda, the buck converter will still have to be there, but with 5 V battery cells in some cases it doesn’t have to be active, or at least it’ll be active for less time during charging.
The point is that you need voltage stabilisation either way — and it doesn’t matter at all whether the voltage to stabilise is 5 V or 4.3 V or whatever. The loss is always the same. (And pretty small with modern converters…)
When can we get samples?
SVOLT belongs to Great Wall Motor, we will see this cobalt-free battery cell in their electric cars first.
The presentation talks about production in H2 2021.
Pedro,
Great content, as always. I have a few questions:
If the BASF HV-Spinel LNMO data is from 2014, why are we only seeing commercial cells now (or in 2021, I guess)?
I saw Dahn’s name on one of the papers you referenced. Is it possible the new Tesla cells are LNMO and not LFP?
What are the disadvantages of the new electrolyte required for high cell voltages?
Do LNMO cells require special thermal considerations?
Why are Panasonic, LG Chem and SDI not jumping on this chemistry, considering it has been around for a while?
Thanks!
Hi Barry.
“So why are LNMO-based batteries not setting new standards for battery performance right now? One of the stumbling blocks is the lack of an electrolyte that can handle the stresses of an LNMO-based battery. Nobody has truly realized the benefits of LNMO cathodes yet because the high voltage it operates at degrades today’s electrolytes and renders the battery useless over time.
However, electrolyte manufacturers are getting very promising results from ongoing research & development that will, at some point, result in electrolytes that will function well in a LNMO battery cell. And when that happens, we are ready with a world-class cathode material that makes optimal use of the benefits associated with LNMO.”
https://blog.topsoe.com/the-cathode-material-for-next-generation-lithium-ion-batteries-is-ready
Here’s what SVOLT says:
https://youtu.be/Amnu5FBL58g?t=2131
Only recently it was possible to have high voltage LNMO battery cells with decent cycle life.
https://ars.els-cdn.com/content/image/1-s2.0-S0378775315303256-fx1_lrg.jpg
I don’t know what additives SVOLT used in the electrolyte. Different additives have different side reactions.
https://aip.scitation.org/doi/pdf/10.1063/1.5005204
https://www.jecst.org/journal/view.php?number=117
LNMO battery cells last longer when cycled at moderate temperatures (25º C).
I really don’t know what is Tesla’s next move, but using LNMO battery cells would be a nice surprise.
I’m pretty sure *all* major battery makers are working on cobalt-free layered oxide cathodes. Tesla has been talking about it for a while now, too. Whether SVolt will actually introduce them first, or is just first to claim they are ready, remains to be seen.
Note that the layered oxide LNMO cathodes are a completely different thing than HV-spinel LNMO cathodes: while both use nickel and manganese, they have them in totally different proportions; have completely different crystal structure; different voltages; different electrolytes…
Both have been researched for a while; and neither has made it to market thus far — though it looks like the layered oxide ones are close now, while HV-spinel is still stuck at research stage AFAIK.
As for “new Tesla cells”, depends on what you mean. It is confirmed that Tesla will use CATL LFP cells in Chinese SR+ in the future; but that’s obviously not the same as what they have been working on in their research, and certainly not a major focus of the elusive “battery day”…
While it’s plausible I guess that they might bring up cobalt-free batteries again, I don’t expect it to be the main focus just yet, considering that all their recent publications about “million mile batteries” talked of NMC and/or NCA…
When they say their new battery “can deliver a 800km of range”, what’s the battery size they’re using for reference?
They didn’t mention it in the video. A 120 kWh battery would probably be enough, considering that the range is in NEDC.
SVOLT seems to have already in production 156Ah cell (i think i read somewhere that it is NMC-811 with Gr./Si anode) which is suited to the VW MEB platform modules. They already offers its own cells based MEB modules. This particular cell is attacking the LG chem LGX E78 pouch cell energy density but with doube capacity, so only 12s1p connection is needed.
By the way it is great to see the latest results of the huge lithium battery research activity around the world particularly in China. These are the results that the EU warned against several years ago. In that time it was not clear for common people (I am quite sure that some do not trust just because they should came from China), but insiders knows how much the EU based research is behind.
On the other side this new technolgies and manufacturers makes even more difficult to be oriented in this area. So again thank you very much Pedro for your great job with your overviews and articles.
These SVOLT cells aren’t NCMA chemistry ? How did you deduce HV ?
NCMA isn’t cobalt-free, SVOLT mentions LNMO in the video.
https://youtu.be/Amnu5FBL58g?t=2152
Another very informative article, thanks Pedro.
If I’m calculating correctly, a 226Ah cell at 4.7V means it’s ~1kwh per cell, so an 80 cell pack, to achieve 400V would be roughly 1.2m x 1m x 0.15m – which seems fairly compact to me. It does seem fairly thick though, so it might take away some foot/leg space for rear passengers in a standard skateboard configuration. For larger vehicles though, this could easily enable 100+kwh packs.
My guess is that any EVs with these cells will need a good TMS that achieves uniform cooling throughout the cells.
Very exciting news indeed.
Voltage is more or less the same as with other layered oxide cathodes (~3.8 V nominal, ~4.3 V max) — i.e. you still need the same 96 cells for 400 V.
Thanks for this Pedro.
The SVolt presentation made good points around need to minimize cobalt, which industry insiders are already well aware of. However, high grade Nickel also needs to ramp up mineral supply volume in the coming 5 years, and sometimes sees disruptive price spikes. Arguably only LFP is truly free from these kinds of mineral supply constraints, but the overall industry will benefit from having wide variety of chemistries / pathways, and some degree of substitutability between minerals, so that no one mineral can become a bottleneck/pain point.
I empathize with SVolt – seeing Tesla, CATL, and BYD getting a large share of the battery headlines recently. If SVolt also have an efficient zero-cobalt approach, and innovative CTP designs, that’s good, and I can understand they want to announce it. They are also emphasizing greater thermal stability, which is sometimes a headlining topic in the China NEV market.
Whilst 4.7V cells should in principle have decent charging, I would like *more detail* than what they have shown – on C-rates
vs. DoD vs. cycle-life vs. temperature. Also a bit more information on their costing – if their chemistry is 10% cheaper than NCx at cell level, that’s great. But I’d ideally like to see a pack level $/kWh costing estimate vs. the other chemistries. Theirs may fall in the gap between LFP (affordable, adequate-,range EVs) and NCx (premium, long-range EVs), but the high voltage characteristic might give it an advantage for some use cases.
Overall – great to see this addition to the existing variety of approaches to better, more affordable EV batteries. This variety lends a valuable *flexibility* and *resilience* to the overall industry.
Thanks Max, you always have interesting insights.
Here’s a simple overview of different cathodes.
https://pushevs.com/wp-content/uploads/sites/6/2020/05/Cathodes-overview-by-Nano-One.jpg
Great – thank you Pedro!
(look out for an email I sent you via the contact address recently).
It makes sense that BYD and CATL are getting more headlines, considering that their new cell-to-pack LFP batteries are going into production now, while SVolt’s new cobalt-free chemistry is still more than a year away… Though of course more exciting, once it does arrive 🙂
Shouldn’t the battery architecture need more focused towards the public transportation model rather than private vehicles?
For humanity’s sake, yes, public transportation makes cities more liveable, and converting to e-buses makes the air much cleaner and reduces emissions more.
However, it is the emotional car buying purchase of private passenger EVs that is going to pay for that. Very few city politicians can raise taxes enough to buy the e-buses needed without getting voted out on the next election, but many people will spend extra money to get an EV (status and environmental consciousness, etc.) It is these people buying EVs that will make batteries cheaper, and when batteries get cheap enough, then e-buses become a justifiable purchase for city managers watching their bottom lines.
Hello Pedro
Can you please tell more about the “HIGH VOLTAGE SPINEL OXYDES”. Is it a hard glassy mineral occurring as octahedral crystals of variable color and consisting chiefly of magnesium and aluminum oxides. Are we talking about negative battery electrode (anode), positive battery electrode (cathode), kind of coating, or kind of electrolyte?
I guess you know the Li-FePO4 chemistry that got fine-tuned during a decade by BYD and Gushen-Godsend :
– 120 Wh/kg at the nominal “0.3 C” rate,
– tolerating 5,000 charge + discharge cycles from 95% to 5% SOC using the nominal “0.3 C” rate,
– tolerating 500 recharge + discharge cycles from 95% to 5% SOC using the fast “3.0 C” rate without cooling,
– tolerating many widely spaced 10 second “6.0 C” discharges without cooling,
– and remaining safe when punctured (please double check this).
Looks fine for a 100% electric car that’s storing 36 kWh (cells weighting 300 kilogram, costing $2,340 FOB)
Looks fine for a serial hybrid car that’s storing 20 kWh (cells weighting 167 kilogram, costing $1,300 FOB).
The hybrid car is viable in case the petrol engine + intercoolers + electric generator + fuel tank + exhaust line weight less than 233 kilogram, and cost less than $1,820 to produce. It is impossible to meet both criterios.
Both cars require a rear-motor “skateboard” architecture kind of VW ID.3 / ID.4 or Tesla Model 3 / Model Y.
Can you please tell if I am right or wrong about the High-voltage Spinel Li-NiMnO4 chemistry that got introduced by BASF and SVOLT :
– 240 Wh/kg at the nominal “0.3 C” rate,
– tolerating 5,000 charge + discharge cycles from 95% to 5% SOC using the nominal “0.3 C” rate,
– tolerating 500 recharge + discharge cycles from 95% to 5% SOC using the fast “3.0 C” rate without cooling,
– tolerating many widely spaced 10 second “6.0 C” discharges without cooling,
– and remaining safe when punctured (please double check this).
Looks fine for a 100% electric car that’s storing 72 kWh (cells weighting 300 kilogram, costing $6,120 FOB)
Looks fine for a serial hybrid car that’s storing 24 kWh (cells weighting 100 kilogram, costing $2,040 FOB).
The hybrid car is viable in case the petrol engine + intercoolers + electric generator + fuel tank + exhaust line weight less than 200 kilogram, and cost less than $4,080 to produce. I think one can meet both criterion using a brand new turbocharged 2-stroke opposed pistons diesel engine that’s delivering 36 kW at 2,200 rpm, feeding two geared variable reluctance electric generators. Both cars require a rear-motor “skateboard” architecture kind of VW ID.3 / ID.4 or Tesla Model 3 / Model Y. Dedicated lightweight direct-driven in-wheel generators 10 kW each, at the front, can take care of the required dust-free electromagnetic braking, and can provide some sort of torque-vectored all-wheel driving.
Could it be we are entering a game-changing zone ?
– new $21,500 Li-FePO4 pure electric vehicles featuring a 165 km range, weighting 1,450 kilogram (100 kW recharge)
– new $25,000 Li-NiMnO4 pure electric vehicles featuring a 345 km range, weighting 1,450 kilogram (200 kW recharge)
– new $25,000 Li-NiMnO4 hybrid vehicles featuring a 105 km range in EV mode, weighting 1,450 kilogram (70 kW recharge)
– the $35,000 Tesla Model 3 Standard Range RWD that’s weighting 1,612 kilogram is doomed, unless it comes standard-equipped with the self-driving feature, along with a per km self-driving subscription say $4.99 per 100 km.
Any kind of reply will be much appreciated.
Have a nice day.
Hi Stephane
High-voltage spinel oxides is a positive electrode (cathode) material.
Regarding cycle life at different C-rates we don’t have data from SVOLT yet. However, some research shows us good results.
https://ars.els-cdn.com/content/image/1-s2.0-S0378775315303256-fx1_lrg.jpg
https://sci-hub.tw/https://www.sciencedirect.com/science/article/abs/pii/S0378775315303256