VSPC develops cobalt-free LFMP battery cell
Lithium Australia subsidiary VSPC developed a new cobalt-free LFMP battery cell. LFMP is the high-voltage version of LFP (LiFePO4) and is expected to become its successor.
Let’s see some highlights of the press release.
VSPC – developing the ‘safe’ lithium-ion battery
- Safety and cost are driving lithium-ion battery (‘LIB’) technology in the direction of lithium ferro phosphate (‘LFP’).
- LFP and its derivatives provide far more duty cycles than competing nickel- and cobalt-based LIBs.
- The addition of manganese to LFP (producing ‘LMFP’) improves LFP energy density while retaining its superior attributes.
- Lithium Australia NL (ASX: LIT) 100%-owned subsidiary VSPC Ltd (‘VSPC’) has successfully produced LMFP cathode powders that demonstrate improved performance.
VSPC has made significant progress towards improving the energy density of LFP LIB cells by adjusting its proprietary manufacturing processes to incorporate manganese into the cathode active material during production.
Improving on LFP performance
VSPC has successfully produced LMFP battery cells for testing. These cells, by virtue of their higher voltage, provide greater energy density than that of standard LFP cells.
The discharge curves below are for cells manufactured from VSPC-produced LFP (left) and VSPC-produced LMFP (right). The higher voltage delivery of the LMFP cells results in an energy density increase of up to 25% when compared with the LFP cells. Globally, major LFP cell producers are striving to achieve similar increases in energy density by introducing manganese as a component of their cathode powder.
LIBs can be divided into a number of categories based on the crystal structure of the cathode materials they contain.
Currently, the types of LIBs most commonly used in electric vehicles (‘EVs’) are nickelcobalt manganese (‘NCM’) and nickel cobalt aluminum (‘NCA’). Both NCM and NCA have a spinel (oxide) structure characterised by relatively low-strength chemical bonds. LFP and LMFP, on the other hand, are composed of phosphates (olivine-like crystal structures) with exceptionally high bond strengths. It is this fundamental physical property that results in the superior characteristics (including thermal stability and long service life) of LFP- and LMFP-type LIBs.
Prelude to rapid- charge batteries for transport applications VSPC plans to develop a rapid-charge battery for transportation applications (see announcement 12 February 2020). Its recent success in testing LMFP cells demonstrates the potential for VSPC’s patented manufacturing process to synthesise LMFP for these applications. Due to its higher energy density, LMFP should reduce the ‘range anxiety’ associated with standard LFP formulations designed for EVs.
Low-cost raw materials
Being able to produce high-performance LIBs without the requirement for nickel or cobalt has many advantages, safety being paramount. Beyond that, the use of common bulk commodities such as manganese, iron and phosphorus reduces costs. Access to much more reliable supply chains is a further advantage.
The ESG advantage
Commercialisation of LMFP for the production of LIBs would eliminate the requirement for materials from regions in which human rights abuse (including the use of child labour) is rife.
Moreover, as noted, using materials derived from industrial waste materials and spent batteries to create precursors for new LFP- or LMFP- type LIBs can enhance sustainability and reduce supply chain risk.
The discharge curves of battery cells with LFP and LFMP cathodes are very different. While LFP battery cells maintain a flat voltage curve from almost full to almost empty, LFMP battery cells have a big voltage drop at around 50 % of SoC (State of Charge).
When battery cells have different discharge curves they also require different BMS (Battery Management System) and GOM (Guess-o-Meter) algorithms. This is something that recently has become quite clear in the Tesla Model 3 SR+ made in China.
The Tesla Model 3 SR+ MIC had its initial NCM 811 battery cells from LG Chem replaced by LFP battery cells from CATL and now the GOM doesn’t give reliable range estimations. Now Tesla needs to gather data from the new cars to make some adjustments to the BMS and GOM, then it can solve this issue with a OTA (Over-the-Air) firmware upgrade.
Anyway, LFMP and LNMO are the two most promising cobalt-free battery technologies for the near future and are expected to become available already next year. They are extremely safe, affordable and have decent energy density. Nonetheless, LFP batteries are getting better and with their proven reliability I expect them to be around for sometime to come.
10% less mAh with ~20% more voltage, not such a big increase after all, while losing low voltage cycle life advantage
One advantage of LFMP battery packs over LFP is that they require less battery cells in series. You can have a 92s1p configuration instead of 110s1p for example, making the battery pack even simpler and more energy dense (less connectors and sensors for example).
Does LFMP have a shorter life cycle than LFP??
According to these old roadmaps from ETC the change from LFP to LFMP won’t change the cycle life.
I don’t know about that… Considering that he voltage is dropping to essentially the same level as traditional LFP half way through, I’m not so sure they can actually reduce the cell count at all.
(It would be a rather marginal improvement anyway.)
Pedro, Very interesting voltage curves. I am surprised that Tesla did not take the time to adjust the GOM algorithm for LFP. Tesla has lots of smart engineers, no? Are there reports of drivers being being stranded by a “suddenly” discharged pack?
BMS and GOM algorithms require a lot of data to be somewhat accurate. I don’t think that this issue is something that Tesla could prevent since only Chinese BMS makers have long experience with LFP batteries.
“An insider said that Tesla suggestion is more about helping the company accumulate and improve data on LFP BMS. Also, this is about North and other low temperature regions, where frequent calibration gives more precise understanding of the battery and prevents from imprecise measurements causing range to suddenly become shorter.”
There’s already a new firmware that adjusts the BMS so that the GOM can provide more accurate range estimations.
Tesla’s model from the begining was that their customers who pay 40k to 100k plus are their beta testers which saves them tons of money and time but is not quite as great for customers who run into issues…
I think this was acceptable in their start up phase as a new company and since BEVs were having their own start up phase for all manufactures…
Tesla is no longer a start up and neirther are BEVs and if they don’t want to permanently develop a bad reputation for quality they need to change their corporate culture which is probably somewhat ingrained already…
Toyota buyers will not put up with second rate quality like early Tesla fans will…
Software issue gave poor range estimates for a couple of weeks… Big deal.
Tesla’s mission requires them to move *fast* — I very much hope they will keep doing that for some time to come.
It appears that more and more Chinese BEV are switching to LFP…
LFP production and installation up and up for November and year…
Ternary up for November but down quite a bit for the year…
Tesla is smart enough to follow science in their business decisions will any other western auto companies??
With covid this year is too early to conclude that ternary production is going down. I think it will still grow in the next years. In Europe (all?) the projected new factories are for NCM or NCA. But yesterday agreement of reducing further car emissions in Europe by 2030 from 40% to 55% adds pressure on the use of cheaper cells and safer supply chains so LFP production in Europe has a new incentive.
Actual specs like volumetric and gravimetric energy density?
Maybe this is so far from production they dont even predict these with acceptable accuracy.
I think that the cycle life isn’t great yet. More data should be released when a stable combination of cathode-electrolyte-anode is found.
They aren’t a battery producer…
Good stuff – the more options we have sourced from relatively abundant and conflict-free minerals, the better.
I have to say, I’m wary of a company in the battery field that calls NMC/NCA (i.e. layered oxides) “spinel” in their official communications…
The 25% capacity improvement seems extremely questionable. The Voltage curve doesn’t suggest anything close to that — and that’s even if the apparently lower mAh didn’t negate some of the gains from voltage… Plus, that’s only the cathode — it’s even less at cell level.
Bringing up conflict minerals is a sleazy PR tactic. Everyone actually caring about the issue should know that the way to deal with conflict minerals is not avoiding the mineral (or the problematic country of origin) altogether, but rather making sure that minerals from that region are sourced responsibly!
Hi Pedro, I actually believe that Tesla has problem with balancing and not really with estimation of energy, even thought those two issues may be connected. Lifepo is quite well known for its inability to balance through its discharge cycle and needs to be charged almost always to 3,6 V to be possible to top balance it. Tesla uses very low power balancers as other manufacturers commonly do, but with lifepo, it may be tricky.
Yes, that makes sense. The flat voltage curve from almost full to almost empty makes balacing pretty much impossible, unless the battery cells are near empty or full. For example, the voltage of a cell with a SOC of 30 % is very similar to a cell with a SOC of 60 %.
If 2020 is the year of “the comeback of LFP”, then 2021 might just become the year of “the revenge of high-nickel”.
There have been several promising messages lately.
Of course Tesla’s battery day and we will probably see the first high nickel/silicon anode 4680 cells next year. But there’s also good news from China.
First of all, an engineer of China’s BEV association recently announced that the industry now understands the thermal runaway problem with high nickel cells. CATL already showed a prototype “only smoke, no fire” NCM811 pack, which will be in production shortly. So that’s kind of a big deal. (Lost the link to the article, sorry).
Now there’s an article about SAIC’s new Zhiji brand, that nicely mentioned the soon-to-be-real new generation cells. Here it is (in Chinese):
It mentions CATL: high-nickel, silicon doped cells, 240 Wh/kg on pack level. Proposed pack: 155 kWh for 1000 km range in the Zhiji.
Also Huineng Technology says to have a high-nickel, silicon-oxide cell. When built in the Tesla long range module box, it should give the car a 1150 km range (Tesla model not specified, probably Model S).
Both companies will bring the cells to mass production in 2021.
So battery development is now accelerating at incredible speed. LFP brings down the cost, high-nickel extends the range. ICE might be dead in the water by 2025.
Indeed, high-nickel content battery cells will be around for a while, at least for premium electric cars.
NCM and NCA will be replaced by NCMA, it combines the best of both chemistries.
Tesla will probably adopt NCMA battery cells made by LG Chem already next year.