Battery Improvements - Pieces of the Puzzle

[new_sneakers]

Slide 34

· We used William’s methods to synthesise these alkali difluorophosphates.

· I overlayed our modern diffraction patterns onto this film – so where the dark rings are represents a strong, intense diffraction, see, they line up just right with the modern diffraction patterns, so we could tell that we were making the right stuff.

· This was invaluable because, to this day, nobody has published the crystal structure of sodium difluorophosphate.

 

[new_sneakers]

Slide 35

· And the ammonium difluorophosphate:

· We took the measurements from Jim Trotter and Harrison, their single crystal measurements, we used them to calculate what the powder diffraction pattern was, and compared them to our measurement – perfect agreement. So those single crystal measurements way back are absolutely perfect. It’s wonderful.

· And with sodium difluorophosphate, we tried to make the single crystals big enough that we could determine the structure, but we failed. We tried with Jason Masuda at Saint Mary’s, but we couldn’t make big enough single crystals.

 

[new_sneakers]

Slide 36

· Anyway the point is, we could make enough of these things and try them in our lithium ion cells.

· So we made enough of sodium difluorophosphate, we call it NaFO, and NaPO2F2 (ammonium difluorophosphate) we called AFO, and methyl ammonium difluorophosphate called MAFO, and I don’t think this [MAFO] was in William’s thesis – we made that ourselves, kinda cool.

· Then we compared to LFO, so we have the cycling data here, capacity versus cycle count, so the lithium difluorophosphate is in blue, and sodium difluorophosphate is in black. Exactly the same – exactly the same. And then if you use ammonium or methyl ammonium, you get worse behaviours, so you’re not going to pick those. But sodium difluorophosphate – exactly the same.

· So we filed a patent on the use of sodium difluorophosphate in lithium ion cells, and [addressing William Reed] your thesis was incredibly useful to us in doing this, it was wonderful.

 

[new_sneakers]

Slide 37

· I just have a couple more slides to talk about.

· Everything I’ve been telling you now has been on this material, NMC532 [second graph down on left side], which has sort of limited energy density, because its specific capacity is only about 170 milliamp hours per gram.

· And in the industry people are moving towards materials like NMC811 [top graph on right], which has larger specific capacity, you can make higher energy density cells with this stuff.

· And it turns out single crystal materials are very important here as well, so we work on single-crystal NMC811 to make long-lifetime cells.

 

[new_sneakers]

Slide 38

· And we recently published a paper, just at the beginning of the new year it came out, where we did cross-sectional SEM studies of the different electrode materials.

· And this is showing NMC811 after 1100 charge-discharge cycles, and you can see [close up picture on right of slide] there is no micro-cracking at all, again.

· Here [top chart above] it’s showing you the charge-discharge cycle life: capacity versus cycle count.

· In blue is NMC 532 single-crystal material, in red is NMC 622 single-crystal which I haven’t talked about and in black is NMC 811 single-crystal.

· When you compare the fractional capacity [middle chart above] of the 811, it’s exactly the same as the 532 – same. Exactly the same.

· And the electrolyte additive in the 811 cells is lithium difluorophosphate [LFO]. It can be replaced by sodium difluorophosphate, and you’re good to go.

 

[new_sneakers]

Slide 39 (final)

· I’ll just make a few concluding remarks.

· Long-life-time lithium ion cells can be made. And I’ve described how to do it.

· For sure, they will last more than 20 years, and power vehicles that do more than a million miles at 100% depth of discharge cycling.

· I would say it’s “easy” to do this.

· What’s most important here is that these cells will be suitable for grid-tied EVs, what I mean there is: you have a vehicle, you drive it to work for example, you park it, you plug it in.

· The utility is talking to it, and it’s talking to the utility. The utility says, oh, we have lots of sun today, let’s charge Jeff’s car. So we charge my car all the way up. Sitting there, not driving, you charge me up.

· I drove home, plug in at home, the sun goes down. The utility talks to Jeff’s car and says, “Jeff only needs to drive 60 kilometres tomorrow, let’s discharge his pack to give him 80 kilometres. That’s enough, for him. So it charges again.

· So here the car is sitting and doing a charge-discharge cycle. If you use a standard lithium-ion cell for that, it’s not going to cut it.

· You need is one of these incredibly good lithium ion cells where you can tolerate thousands of charge-discharge cycles, full depth of discharge, then you can grid-tie, and become part of the energy storage solution.

· For me, if you have a vehicle and you’re doing that kind of thing, you should be paid by the utilities for storing and delivering energy back.

· I think the future’s pretty exciting with this kind of technology.

· The final comment I want to make is: I really want to emphasise the importance of basic research like William Reed’s. Who knew in 1965 that what he did was probably – I mean would it have been useful, useful? I don’t know, from a practical point of view.

· Thank goodness the work was done and it was well-preserved here in the UBC library – that’s another incredible thing. That the thing was copied with those X-Ray patterns taped on to the piece of paper.

· I think it’s really likely that the sodium difluorophosphate can become a replacement for lithium difluorophosphate, and will help drive down the cost of lithium ion cells.

· So [addressing William Reed] thank you very much for your work, and that’s all I have to say. Thank you.

· [Applause]

 

[MC3OZ]

I think it’s really likely that the sodium difluorophosphate can become a replacement for lithium difluorophosphate, and will help drive down the cost of lithium ion cells.

This is an important point that some in the media might be over looking, or don't currently understand..

There are lots of opportunities for lowering the costs of batteries and improving longevity.

Tesla and Dahn seem to have tackled most of the relevant areas where improvements can be made.

 

[Robertj]

Thank you very much for posting this

I wonder why the original video was taken down ?

I am sure we will find out more in September

 

[new_sneakers]

One takeaway is that the million-mile battery only represents two changes, and Tesla is doing much more.

It’s worth checking in on @ReflexFunds post from January, linked earlier in this thread, I’ve bolded the two changes.

***
Thoughts on what Tesla will announce on battery day in April & how Tesla’s future battery strategy will come together:

• Use cell supply from Panasonic/LG/CATL to bridge to ramp of in-house cell production (possibly towards ~90GWh contracted from these three suppliers).
• Announce that in-house cell production has just started (Apr-20) on a small scale (likely for Semi or Plaid Model S), with plans to ramp significantly in 2021 (potentially for all future new capacity from 2021).
• Announce a roadmap to reach 2TWh of annual in-house battery cell+module+pack production capacity by 2030. Enough for ~20 million annual EV sales and ~750GWH annual stationary battery storage sales.

Possible relatively short term technology breakthroughs:

• Tesla will apply agile development to its in-house cell manufacturing as it does everything else - so flexibility for rapid upgrades and iterations of the process to accelerate cost experience curves.
• Use Maxwell dry electrode tech to reduce manufacturing cost and footprint.
• Maxwell dry electrode tech leads to better physical properties, in particular allowing thicker cathodes (higher cathode % per cell) & possibly new chemistries.
• Move to use of single crystal cathodes - possibly helped by Maxwell process/other in-house R&D. This was a big part of the 1 million mile cells tested by Dahn.
• Use very carefully selected electrolyte additives following Dahn research.
• Highly automated manufacturing process to reduce staffing bottlenecks to production ramp.
• Tesla Hibar designs systems for electrolyte insertion during the cell manufacturing process.
• Combine all this with further in-house developed cell IP and possibly third party licensed tech. (Remember there are many steps in cell manufacturing and Maxwell/Hibar are only part of this)
• Reduce cathode kg per kWh to reduce raw material cost
• Next generation in-house module/pack lines for continued reduced cost & capex.
• Build a huge factory to build in-house cell/pack manufacturing equipment at scale (the machine that builds the machine that builds the machine) - significantly reducing capex per GWh capacity

Possible Longer term breakthroughs:

• Integrated cell & pack design & manufacturing process to reduce footprint & cost.
• Dahn lithium metal anode allows for much thinner anode, higher energy density & longer electrode life (at the expense of shorter electrolyte life).
• Replaceable electrolyte design to extend lithium metal anode battery life. Develop Hibar machines for easy electrolyte replacement in service centres.
• Dahn research is used to eliminate cobalt from the cathode leaving just Nickel Aluminium or Nickel Manganese.

Note these are all just possibilities (based on acquisitions, press leaks, published scientific papers, patents & speculation):
These various steps & incremental improvements may or may not be introduced once they have been proven ready for affordable mass manufacturing.

Some things I thing would help accelerate and de-risk Tesla’s battery cell ramp plans:

• Buy Panasonic’s GF1 business for cell manufacturing employee experience (who can be used to train new employees on Tesla’s cell lines) and other cell IP.
• Buy/build Cathode powder manufacturing expertise (currently Panasonic mostly uses Sumitomo). Cathode powder is likely ~20% premium to its raw material constituents & the process can be key to cell properties.
• Buy Nickel Sulphate & lithium carbonate/hydroxide processor expertise - this will be a huge % of cell cost & Tesla’s plans require ~10x the current Nickel sulphate & Lithium market size.
• Buy other suppliers in the cell manufacturing chain

Tesla cannot trust & rely on third parties to deliver such critical components of its business plan, particularly when the metals market leaders do not believe in an EV transition as aggressive as Tesla is targeting.

***

 

[S3XY]

I actually grabbed a copy of this video back when it came out.

 

[Baumisch]

Thank you for this great summary of a video I couldn't make myself watch as it was tooo ... academical presented ... or so ...

2 Takeaways for me personally:

- This sounds like 2 changes are needed, single crystal and some additives - sounds like an easy change for existing 18650 in the S/X Packs
- I can see Tesla operating one line of 18650 cell manufacturing to produce replacement packs for older S/X while keeping some margin and staying below my pain-point (I'd say 10-15% of new S/X is my threshold for replacement costs - so 8-12k€ in my case).

 

[Ed Hart]

We have a date! Finding it hard to wait!!