Developments on LFP battery technology

SVOLT battery roadmap
SVOLT battery roadmap

Nowadays, good electric cars already have enough range for most people, but they are still much more expensive than their ICE (Internal Combustion Engine) counterparts. This is why LFP (LiFePO4) and CTP (cell-to-pack) are extremely important technologies to make electric cars mainstream. Automakers that don’t plan to use these two technologies as soon as possible aren’t serious about mass producing electric cars. For example, Stellantis plans to start using CTP packs with LFP cells only by 2024


LFP is a cobalt-free battery chemistry that combined with simple CTP batteries can finally make electric cars compete with ICE cars on price and availability.

While at the cell level the energy density isn’t great, at the battery pack level LFP can compete with other chemistries. Since LFP is a very safe battery chemistry and cells don’t burn or explode even if punctured, battery packs don’t require much protective equipment. Therefore, LFP battery packs are extremely simple to assemble and can adopt a module-less CTP configuration.

As for common NCA and NCM cells, they are more energy dense, but aren’t very safe. Battery packs made with these cells require modules and metal plates to act as firewalls in the case that a cell burns or explodes.


Summing up, with super safe LFP battery packs the VCTP (volumetric cell-to-pack) and GCTP (gravimetric cell-to-pack) ratios are much higher. Let’s see some average figures.


LFP battery packs

  • VCTP ratio: 60 %
  • GCTP ratio: 85-90 %


NCM/NCA battery packs

  • VCTP ratio: 40-45 %
  • GCTP ratio: 60-65 %


The VCTP ratio tells us how much of the battery pack’s volume corresponds to active material – that actually stores energy (cells). The rest of the volume is from the passive material used to assemble and protect the cells (case, modules, cables, sensors, BMS, TMS, etc).

The GCTP ratio tells us how much of the battery pack’s weight corresponds to active material – that actually stores energy (cells). The rest of the weight is from the passive material used to assemble and protect the cells (case, modules, cables, sensors, BMS, TMS, etc).


As you can see, not only the NCA and NCM cells are by themselves more expensive than LFP, their battery packs are also much more complex and require expensive material to make them somewhat safe. Only around 45 % of the volume is used by active material (cells), which means that the passive material required to assemble and protect the cells takes most of the space.


Below you can see the simplicity that BYD achieved in 2020 by removing modules with the introduction of its Blade battery that follows a CTP configuration.


BYD battery pack evolution

BYD battery pack evolution



Moving on, let’s see what kind of energy densities important battery cell makers expect to soon achieve with LFP battery cells.



  • 2021: 170 Wh/kg (graphite anode)
  • 2022: 200 Wh/kg (graphite anode)
  • 2023: 230 Wh/kg (hybrid graphite/silicon anode)


SVOLT expects to increase the energy density of LFP cells by adding more silicon to the graphite anodes.



  • 2021: 230 Wh/kg (207 Wh/kg at pack level with JTM)
  • 2022: 260 Wh/kg (234 Wh/kg at pack level with JTM)


Guoxuan expects to increase the energy density of LFP cells by replacing graphite with silicon in the anodes.



  • 2021-2023: 180-200 Wh/kg (350-450 Wh/L)
  • 2023: 210-230 Wh/kg (450-500 Wh/L)


By 2023 CATL expects to introduce the LxFP battery chemistry, which is probably the high-voltage version of LFP (LMFP/LFMP) that I have been writing about for some years.


CATL battery roadmap

CATL battery roadmap


By now you probably know that the BYD Blade battery is my favorite battery pack design. I cringe every time I watch a video of Sandy Munro tear down a battery pack from legacy automakers. There is so much junk in there that could be avoided with a simple CTP battery made with LFP cells. Imagine how simple and fast can be the production lines that assemble CTP batteries.


When first released in 2020, the BYD Blade battery achieved an energy density of 166 Wh/kg at the cell level and 140 Wh/kg at the pack level. However, LFP chemistry has been improving since then and I wonder how energy dense will be the second generation. If BYD reaches 200 Wh/kg at the cell level, the Blade battery pack can reach 170-180 Wh/kg.

I’ll be disappointed if by next year BYD doesn’t use silicon as anodes for faster charging and reach at least 170 Wh/kg at the pack level.


The imminent arrival of BYD e-platform 3.0 is a good opportunity to introduce the second generation of Blade battery. I’m curious to know the energy density of the battery pack used in the upcoming BYD Dolphin.


BYD e-Platform 3.0

BYD e-Platform 3.0

Pedro Lima

My interest in electric transportation is mostly political. Iโ€™m tired of coups and wars for oil. My expectation is that the adoption of electric transportation will be a factor for peace and democracy all over the world.

21 Responses

    • Mike/Liverpool says:

      Sodium ion is a game changer, China will MASS produce these all over the World.
      Everyone will have a “Power wall “…Wind blows at night?… problem just route it to everyones Power wall.

      It will become vastly cheaper than any other…..& its already at 130-140 wh/kg

      • Pedro Lima says:

        Thanks Mike.

        If SIBs (sodium-ion batteries) finally reach 200 Wh/kg at cell level as expected, with CTP packs they can even be used in EVs.

      • Mike/Liverpool says:

        CATL can make 600Wh/kg battery?!

        • Pedro Lima says:

          Those are probably solid-state lithium-metal batteries, they are still many years away.

          • Mike/Liverpool says:

            Once they do the EV will be lighter than an ICE version

  1. Seowoo Lee says:

    It is great article! I appriciate your effort on this type of research.

    Here are some arguments that I do not agree with. Please, let me know if I am wrong.

    volumetric and gravimetric cell to pack energy densities also depends on the battery’s form-factor (casing such as pouch, cylindrical, and prismatic).

    • There is not a clear condition how you made the comparison between LFP and Hi-Nis. You are assuming LFP is prismatic and NMC or NCA is not prismatic. When you make a comparison, it should be ‘Apple to apple’ not ‘apple to orange’.

    If you believe LFP is better than Hi-Ni type of batteries for integrating into EVs, why there is not a single performance EV that equipped with LFP batteries?

    • It is fact that LG or SKI’s Hi-Ni battery integration is having hard time to do the CTP due to safety concerns such as thermal management.
    • It is true that LFP has a better safety aspect than Hi-Nis
    • It is true that the theoretical energy density of LFP cannot be better Hi-Nis
    • In the same form-factor LFP and NMC or NCA in the same volume of pack, LFP never can be better than NMC or NCA.
    • This is why LFP and Hi-Ni batteries have their own applications.

    ‘SVOLT expects to increase the energy density of LFP cells by adding more silicon to the graphite anode’ and ‘Guoxuan expects to increase the energy density of LFP cells by replacing graphite with silicon in the anodes’

    • Unfortunately, this is false. ‘the energy density’ comes from cathode active materials not anode active material. Even though you increase amount of anode active materials, it does not improve overall the energy density of the battery.
    • Silicon materials in anode may help fasten lithiation process (improving fast-charging capability, capacity retention, etc.) but, it does not help to improve the energy density.

    Hope to see your another good article soon!

    • Pedro Lima says:

      Hi Seowoo, welcome.

      Let me explain.

      If the inside material is the same, pouch cells are always more energy dense than prismatic. If you use modules you better also use pouch cells. You don’t gain anything by using prismatic cells inside modules.

      The advantage of prismatic cells is that you can avoid modules, if the cells are safe enough.

      Patentsโ€™ restrictions over LFP cathode chemistry will start to expire in 2022, then cell makers will be able to produce LFP cells outside China without paying royalties. (check point number 3)

      I’m pretty confident that next year CATL will start producing LFP cells in Germany to supply at least Tesla.

      As for the silicon anodes, you are partly correct. Anodes are more important to determine power density, but they can also slightly increase the cell’s overall energy density. I think that Tesla already explained this on Battery Day. But if you want I can show you some research articles on this topic.

      By the way, check out my article with a comparison of different batteries in 2020. You’ll notice that where it matters – at the pack level -, LFP can already compete with those.

      Feel free to point anything in my articles that is wrong or poorly explained. I also learn from comments made here.

  2. Famlin says:

    Professor Pedro: Crystal clear presentation. Now I learnt about volumetric and gravimetric ratio.
    Few years ago, bp in its world energy stats started publishing details about prices of Cobalt and Lithium carbonate and everyone used this data to scare that rising cobalt prices will block the EV movement. Advanced NCM with lesser cobalt combined with the LFP has put a stop on Cobalt price increase.

    2017: $55.79/ton
    2018: $72.79/ton
    2019: $33.20/ton
    2020: $31.44/ton

    It increased in 2018, but then dropped from 2019.
    2020 could be the year that LFP hits parity with NCM on 1:1 basis.

  3. Famlin says:

    @Pedro: 1 more question. Do you think even the pack can be bypassed with the cell to structure where the prismatic LFP can be placed directly in the structure of the vehicle. Will be volumetric and gravimetric ratio of such a system will be even better.

  4. yoyo says:

    Great article…
    Never knew China got a free ride on LFP IP and everyone else pays??
    2022 tic tic…

  5. Lambda says:

    Thanks for the article, I really enjoyed it!

    Any insights on Blackstone Resources AG’s LFP cells with their thick layer technology? Are they not subject to LFP patent restrictions? Do you think they might be a reasonably important player in the field given that they also have mining rights in South America?

    Any battery manufacturer talking about lithium metal anode LFP cells in the 2025 time frame or so?

    And why do you think Tesla and others are still holding out with the cylindrical cell format given the geometrical limitations and the need for new tabless technologies? Is it inertia or is the sandwich pack structure easier to achieve with cylindrical rather than prismatic cells?

    • Mike/Liverpool says:

      Meantime over at VW
      Volkswagen Group Outlines NEW AUTO Strategy Through 2030 (

      ICE is not going any time soon…….in fact i bet H2 engines will arrive soon after 2030

      • Lambda says:

        Had VW bought Lucid, they would have a platform which doesn’t seem too different from the SSP some 5 years earlier than planned, especially if they had gotten the cell-to-pack technology right. Would have needed modification to reach entry-level cars, but they would not be playing catch-up with BYD’s Dolphin. Right now, a startup like Lucid will be producing a car by the end of the year that will likely be equivalent to Audi’s Artemis coming in 2024/2025. Not good.

    • Pedro Lima says:

      Thanks, I’m glad you enjoyed it.

      I really don’t know what to think about that company Blackstone Technology. Its website is all about getting investors, without technical data on the LFP cell.

      I’ve seen some research papers about lithium metal anodes combined with LFP cathodes, but don’t know any battery cell maker that has them in its roadmap.

      The cylindrical cells have the advantage of fast and simple production, maybe that’s why Tesla still bets on them. You can wind a cylindrical cell extremely fast by using simple machinery.

      Nonetheless, I still prefer module-less packs made with prismatic cells. I think that the simplicity of the BYD Blade battery is really the way forward to build cheaper and reliable packs.

      As Sandy Munro says: simplify, simplify, simplify!

      • Lambda says:

        Thanks again for the insights! Agreed about simplicity. I imagine a REE Auto platform with an advanced cell-to-pack LFP battery pack would be quite a sight to behold. About the stacking, I quite liked the comparison to stacking money that VW made in its Power Day presentation:

        Might be hard to do with long Blade battery cells, but I don’t see why you can’t have layers being deposited on top of each other like a commercial printer layers paper after printing.

        By the way, iM3NY’s LFP cell manufacturing might be a worthwhile addition to the list above. Seems like a mix of LFP and LMP, some limited technical data and roadmaps are given, and they purchased equipment from A123 Systems:

        Really looking forward to the BYD Dolphin coverage on this site!

  6. Marcel says:

    Thanks for the nice walk through Pedro. From Seeweo Lee’s comments, I didn’t realize that there were China specific patents on LFP technology. That’s very interesting. If the patents end in 2022, then hopefully we’ll see a massive global expansion of LFP manufacturing after that.

Leave a Reply to Lambda Cancel reply

Your email address will not be published. Required fields are marked *