Tesla: Strategic Battery Strategy (NASDAQ:TSLA)

Tesla (NASDAQ:TSLA) will likely make 1.5+ million electric cars and SUVs in this coming year having sold just shy of a million vehicles last year. The company says it will make 20 million electric vehicles a year by 2030. How many cars is that? Well, let’s look at the world car market.

Top Carmaker Worldwide Sales in 2021

Tesla isn’t on this list of largest carmakers today. But, by 2030, Tesla will sell more cars, trucks and SUVs than the two largest carmakers, Toyota and Volkswagen, combined. And, those 20 million Tesla sales will be coming from others’ market shares. Think about that.

Legacy carmakers make almost entirely internal combustion engined vehicles today. By 2030, roughly half of new cars sold are expected to be electric. This also means that, by 2030, the ICE car market will be about half of what it is today, the other half of the market being BEVs. Just to keep existing market share, a legacy ICE carmaker will need to switch half of its production to BEVs by 2030.

The race is on to see which carmakers will transition quickly enough to retain or expand current market share. Some will. Others won’t. And those that don’t will lose economies of scale and likely end up in bankruptcy court.

Availability of electric vehicle batteries is the key factor that will allow some manufacturers to transition quickly and keep others from doing so. Nobody is going to buy an electric car without a battery to make it go. Battery supply will be limited primarily by the availability of key battery feed stocks, particularly class-1 nickel and battery grade lithium carbonate and hydroxide. The earth contains huge amounts of nickel and lithium – far more than needed for batteries. The hard part is going to be ramping up the mining and processing supply chains fast enough to meet EV battery demand.

The war in Ukraine and associated trade disruption have crimped supplies of class-1 nickel used in electric vehicle lithium batteries. Recent COVID lockdowns in China are similarly crimping supplies of battery grade lithium carbonate and hydroxide. All the while demand for battery grade nickel and lithium feedstocks is soaring as automakers rush to transition to BEVs before competitors. It follows that using battery designs that deliver the most kWh, and hence, the most cars per ton of feedstock can prove strategically important as car companies fight to hold/grow market share during the rapid BEV transition. In terms of company survival, the ability of a particular cell type (chemistry) to deliver more cars will relegate battery cost to a secondary consideration.

There are three general classes of lithium battery that look to be significant for electric vehicles going forward: Iron phosphate – LFP, high nickel – NCM/NCA/NCMA, and high manganese – LNMO. Each requires different amounts of lithium and nickel per kWh of cells. Choosing a cell chemistry that delivers more kWh of batteries and hence more electric cars per tonne of nickel or lithium feedstock can allow a carmaker to deliver more vehicles and hence better defend/grow market share under conditions of feedstock scarcity.

Newer cathode chemistries that eliminate cobalt and optimize energy density will drive feedstock demand going forward. We need to look at where designers are heading, for instance, high nickel (90% Ni aka N90) rather than NCM111 and to cells with silicon dominant anodes because this anode choice both increases energy density compared to graphite anode and allows high charge rates. (Lithium metal anodes are also being looked at, but I believe these will not become commercially viable for some time. Cobalt while expensive and limited in supply is being designed out of batteries and likely will not impact new battery supply mid-term and on.)

Different cathode chemistries result in different cell voltages and the higher the cell operating voltage, the greater the amount of energy stored in each lithium ion passing between anode and cathode. More energy per lithium ion means less lithium to store each kWh. Also, fewer lithium ions mean less cathode material (nickel, iron, cobalt, manganese) and less anode (carbon, silicon).

Additionally, some lithium is required for the electrolyte and to form the SEI that protects against undesirable side reactions. Roughly speaking, the amount of lithium needed for the SEI is proportional to the amount of anode material which in turn is proportional to the amount of lithium ions shuffling back and forth to store energy.

Lastly, in the case of high nickel cathodes, a substantial amount of excess lithium must remain in the cathode to support the cathode crystal structure. Remove too much of the lithium from a high nickel cathode (i.e. overcharge the battery) and the layered crystal structure will collapse resulting in lost battery capacity. This increases the amount of lithium required. It also increases the amount of nickel (and other cathode metals) required per kWh. About 30% excess lithium is typically required to achieve acceptable cycle life with high nickel cathodes.

The following table shows the amounts of lithium and nickel (in g/kWh) required for a cell with each cathode chemistry. [Assumptions: Silicon dominant anode; electrolyte + SEI require 5% of active lithium amount; N90 requires excess lithium equal to 30% of the combined active, SEI and electrolyte lithium.]

This data can be used to compute how many 60 kWh battery packs can be made from each tonne of available feedstock. For simplicity, I have assumed zero manufacturing loss/scrap. If a manufacturer is building cars similar to Tesla’s Model Y, the number of 60 kWh packs is a rough measure of the number of such cars that can be made.

In the case of nickel, we first observe that LFP requires none and is therefore not constrained by nickel availability. High nickel N90 and LNMO cathode do use nickel and are constrained by nickel availability. The following chart illustrates the very large production volume advantage LNMO offers over N90 under constrained nickel supply. Since LNMO and N90 offer similarly high energy density choosing LNMO cathode for longer range, higher performance vehicles offers significant advantage. A manufacturer using LNMO rather than a high nickel cathode chemistry can deliver more than three times the cars from the same nickel supply.

When it comes to lithium, all three cathode types are impacted by lithium availability. Again, LNMO offers maximum production volume under constrained supply. A manufacturer using LNMO cathode will be able to deliver more than three vehicles for every two vehicles that could be made with LFP and nearly twice the vehicles compared to using high nickel cathode for a given supply of lithium.

Four carmakers and probably more are looking to use LNMO cathode. Volkswagen, Tesla, Stellantis and Great Wall (OTCPK:GWLLF) are named in this article. In a recent episode of “The Limiting Factor”, Jordan Giesige corralled Elon Musk to the point of indicating that Tesla’s manganese cathode will indeed be LNMO.

Is LNMO Ready For Prime Time?

LNMO, also known as HVS (High Voltage Spinel) has been around as a lithium battery cathode material for many years but has not been successfully applied. HVS has had two major drawbacks. Early versions of HVS suffered capacity loss during cycling because the crystal structure rearranged in a way that reduced the battery voltage from 4.75 Volts (vs. Li/Li+) down to roughly 4 Volts within a few tens of cycles. Development of cathode processing that delivers precisely uniform crystal structure combined with addition of dopants (Sn, Cr, Nb, etc.) has produced HVS that retains crystal structure and voltage over many cycles as demanded by electric vehicle applications.

The second problem results from the high operating voltage of HVS cathode. High voltage accelerates manganese leaching into the electrolyte which in turn supports parasitic reactions with the battery anode, gas generation in the cell and loss of capacity. Initially, HVS was thought suitable only for solid state batteries where a solid electrolyte would surround cathode particles and minimize manganese leaching. Nano One (OTCPK:NNOMF) has worked for several years with Volkswagen on development of LNMO/HVS for such solid state batteries.

Disclaimer: Nano One is a small essentially pre-revenue company. Investors should seek professional advice, perform thoughtful due diligence, and limit any investment in this company to funds they are prepared to put fully at risk. I am long NNOMF. That alone does not mean you should be too.

It turns out manganese leaching when HVS is used with conventional liquid electrolytes can be prevented by appropriate coatings. A year and a half ago, Nano One announced that cells with LNMO/HVS coated and stabilized cathode from their “One-Pot Process” using conventional carbon anode and conventional liquid electrolyte have achieved 1,000 charge/discharge cycles at room temperature and 500 cycles at 45C. This level of cycle life is sufficient for electric vehicles and was achieved without electrolyte optimization and other “tweaks” likely to be implemented in a commercialized battery design.

Several cathode material suppliers (Haldor Topsoe, NEI Corporation, and Targray) are now offering LNMO cathode material commercially as described in this article. These cathode suppliers are all closely held. Targray is a distributor. NEI Corporation supports nano tech research. Only Haldor-Topsoe is aiming for volume LNMO production, planning 100 tonnes per year capacity by 2023.

Cathode process developer Nano One is public but does not offer LNMO or other cathode materials commercially. I put the question of whether the company is or has been supplying LNMO to any battery makers or electric vehicle OEMs to CEO Dan Blondal and got the following reply:

We do sample LNMO and typically in kilogram quantities. That is sufficient for most parties to make prototype cells that are representative of large commercial cells. We have a handful of active evaluations and collaborations underway with cell developers / mfgs and EV OEMs, testing our LNMO materials in their HV battery systems.

A look through Nano One’s press releases offers a look at the development timeline for LNMO.

This snapshot of LNMO development shows initial fixes for LNMO roadblocks – crystal instability and manganese leaching – through EV OEM evaluations and engineering studies for industrial scale production. All of this in just four years’ time.

Note that I have used Nano One announcements here because they layout a more or less continuous view. There are other players, particularly Haldor-Topsoe and potentially Albemarle (ALB) using a Battelle developed process. And of course, Tesla with its in-house cathode material processing displayed at Battery Day.

Appreciate that these timeline announcements are dated after the fact and the actual state of play is almost certainly further ahead. Understand as well that it is the timing of the first mover that will matter as companies strive to defend/grow market share. By the time most of the industry has switched to LNMO, the first movers will likely have already pressed the strategic advantage offered by this cathode chemistry to expand market share in BEVs at competitors’ expense.

Likely First Movers

At this point, being “first mover” is not about inventing LNMO as a new cathode, researching whether it can be used in conventional electrolyte cells or figuring out a process to make LNMO. All of those things are already done. The next and the critical move is to make and use LNMO at scale. Only when in use at scale do the strategic advantages of LNMO described above come into play. First movers will be those who get LNMO batteries into their cars, at scale before their competitors.

Tesla with its vertically integrated battery manufacturing, including in-house cathode processing is, I believe, best positioned to move rapidly to LNMO cathode probably for use in its 4680 cells. The company has repeatedly talked about their constant quests for nickel and lithium supplies and implementing LMNO to displace high nickel and even some LFP batteries in their vehicles would offer less constrained growth for this BEV industry leader.

Volkswagen is my follow-up. VW seems to be betting the company on BEVs and needs to expand vehicle production and its suppliers’ and/or in-house battery cell efforts accordingly. VW also has several years’ experience with LNMO development while working with Nano One. Ford, via its partnering with VW could well be brought along to using LNMO if/when VW takes that step.

Switching from high nickel or LFP to LNMO is a “system level” optimization. LNMO batteries offer almost as much capacity as high nickel batteries and cost only a bit more than LFP batteries. And LNMO has a flat discharge characteristic (like LFP) and a higher operating voltage than even high nickel batteries, requiring redesign of a vehicle’s battery configuration and battery management electronics.

This makes LNMO more of a compromise choice than a world beater choice on a cost/performance basis. The area where LNMO stands out is in positioning a carmaker to more robustly compete for market share when feedstock supplies are scarce. Only a carmaker viewing their strategic business objectives and having insight and control of their deep supply chain will be likely to both appreciate LNMO and take full advantage of it.

A carmaker relying on an outside battery maker would have to be “sold” by the battery supplier and even then be dependent on the battery supplier’s supply chain to fully realize LNMO’s strategic advantage. Small carmakers will likely find themselves in this situation and also be less able to expend the engineering resources to switch to LNMO from LFP and high nickel batteries on offer in the market place.

Conclusions

The auto industry is undergoing a steep, S-curve transition from ICE to battery-electric propulsion. Both new entrants and established legacy manufacturers are racing to make more BEVs in order to expand/defend market shares. At the same time, battery supply is constrained by availability of key battery feedstocks, particularly class-1 nickel and lithium carbonate/hydroxide as supply chains try to keep pace.

A long studied high manganese cathode material now ready for use in EV batteries will allow dramatically more batteries and hence more cars to be made from limited supplies of nickel and lithium. This LNMO cathode competes with high nickel in performance and with LFP on cost. Multiple EV OEMs are evaluating this cathode material, and multiple suppliers are working toward volume production of LNMO.

OEMs switching from LFP and high nickel batteries to LNMO will be able to make dramatically more EVs for given supplies of lithium and nickel. LNMO offers strategic advantage to OEMs in the race to make more EVs and expand/defend their market shares. The ability to make more EVs under conditions of feedstock scarcity will prove to be a key element separating winners and losers in the ICE to electric transition.

Tesla and Volkswagen appear to be well positioned to move to LNMO batteries at scale. Both companies have included high manganese cathode in their battery roadmaps. Tesla has confirmed that LNMO will be their high manganese cathode choice. VW has worked for several years with cathode process developer Nano One on LNMO cathode material.

In the case of Tesla, switching to LNMO for some/all of its long range vehicles that use high nickel batteries could stretch nickel supplies enough to support volume Cybertruck and Semi production.

Other OEMs may also adopt LNMO cathode via commercial cell suppliers though the “sell cycle” may be more protracted.

Investors should pay close attention to which OEMs move to LNMO because those that do will gain substantial production volume advantage and be better positioned to defend/expand market share as the electric vehicle S-curve disruption continues.