Mpac – Battery cell manufacture

In my previous article I looked at Mpac from the perspective of information published by them, including a run down of all the things that could go wrong with their contract with FREYR. Here I reveal the fruits of further research into that contract.

Sources of information

The other parties to the contract are FREYR, but also indirectly 24M from whom FREYR licence their technology. And it appears that 24M were instrumental in Mpac getting the contract – from the original announcement:

Mpac was prequalified to participate in the competitive tender following nearly three years of cooperation with 24M Technologies (“24M”) on industrializing and scaling 24M’s SemiSolid lithium-ion battery platform technology.

This opens up the possibility that Mpac might win business from other 24M clients. But to reiterate, the baseline position for an investor should be that 24M will fail just like most of novel battery technologies will.

Anyway, we have two obvious venues of research:

• FREYR Battery: Home Page, News, Investor Relations, SEC Filings, YouTube, Stockopedia
• 24M: Home Page, Press Releases, SEC Filings, YouTube

FREYR are generally more helpful as they are a listed company, potentially in search of further future public equity funding. On the other hand, 24M are a private company co-owned by private equity and a variety of industry partners who have actively been removing information from their website and don’t even link to their own press releases.

Along with the above sources there are the usual trade magazines, partner press releases, government funding websites etc. In terms of 24M technology this is the best description I can find, although there is also some detail on archived versions of 24M’s website.

FREYR / Mpac Contract

FREYR’s S-1 Prospectus explicitly mentions the contract with Mpac, leaving no doubt that they are sole supplier of casting & unit cell assembly machinery, and that this is not just an important part of the production process, but lies at its very heart. It then goes on to give the contract price of $13.8m and the phasing of payments.

But more information still is to be found in the (only lightly redacted) copy of of the FREYR / Mpac contract included as an attachment.

Revenue to date

From this I have determined that Mpac was paid £4.0m on signing of the contract (23/7/2021) and another £4.1m was due on commencement of equipment build. I have many reasons to believe that the second as well as the first payment was both received and recognised as revenue in H2 2021:

• The milestone date given in the contract was 11/11/2021
• I believe “building the equipment” would be done in the UK and so there is no external factor such as FREYR’s building fit-out that would delay it
• Payment was confirmed as paid in FREYR’s Annual Report, albeit they do not explicit say it was paid by 31/12/2021 rather than by 9/3/2022 when the AR was signed
• In the post year-end trading statement Mpac said the contract remained on track
• There is an otherwise unexplained jump in Mpac’s FY 2021 revenue not from Pharma, Healthcare or Food & Beverage in the segmental analysis note to the annual report and this matches the expected amount
• My reading of the contract is that satisfaction of the performance obligation coincides with payment
• There was no increase in Mpac’s contract liabilities in 2021.

Of course, the downside to this is that the underlying 2021 figures weren’t as strong as they appeared. Revenue from the core pre-Lambert, pre-Switchback, ex-FREYR business was down circa 20% YoY according to my calculations. However, business here is inherently lumpy and excluding a customer just because it is new isn’t really fair. And it is especially not fair when there is scope for significant further revenue in future.

Future CQP revenues

With 80% of the $13.8m contract size already paid, the remainder ultimately due from the Customer Qualification Plant (CQP) is necessarily modest. The original milestone dates were 23/6/2022 for the Factory Acceptance Test (5%) and 19/9/2022 for commissioning (15%). It is clear however that the build on FREYR’s end has suffered issues, with start of production now delayed until “late 2022” according to FREYR and Mpac revising it from Q3 2022 at the January trading statement to Q4 2022 by the time their Annual Report was prepared in March.

While Mpac might now be able to claim extra costs for the delay, the strength of the dollar should work in their favour and these costs should be minimal even if they pursue this. The question is whether the final 15% payment will make it into FY 2022 and I have to say on the balance of probabilities it will not, either due to further delays at FREYR or last-minute hitches with Mpac’s equipment.

Future Gigafactory revenues

The FREYR / MPac contract also includes options for much larger Gigafactories, with a firm price agreed for Plant 1 and a benchmark price for Plant 2 & 3. Initial deadlines for “execution” are the fourth quarters of 2022, 2023 and 2024 respectively. Unfortunately these contract prices are redacted.

Such is the indicative demand for batteries that FREYR’s current plan is to combine Plant 1 and 2 together into Gigafactory 1&2 under a single build with a total nominal production capacity of 13.25GWh/year, the exact figure dependent on the precise battery chemistry / mix demanded, using a building the area of 7ha and 8 production lines.

This first production plant (or Gigafactory, if you must) compares to 375MWh/year, a covered area of 1.3ha and one line for the CQP (aka Pilot). So, depending on the way you look at things, Plant 1&2 is 35x, 5.4x or 8x the size of the CQP. Some of this disparity can be explained by scale advantages of larger factories and greater line speed enabled by the faster machines specified in the option contracts. But in broad terms the contract size is likely to be in the order of 10x the CQP. Assuming a similar milestone structure to the CQP then the 80% pre-payments alone would double Mpac’s revenue at a stroke.

It is clear from FREYR’s Q1 presentation that the final go-ahead for production plants is dependent on converting their conditional offtake agreements to definitive sales agreements so they are on the path to receive project financing. Here they claim 50GWh of demand close to definitive, 100GWh total conditional, a further 100GWh under negotiation and 200GWh/year under discussion.

But while customers are fairly desperate for batteries right now, it was initially unclear whether they would make a definitive purchase agreement without any samples from the CQP, which I understand is running 2-3 months late. Although FREYR declined to address this as a question to their AGM, Investor Relations (IR) have indicated via email that the CQP does not need to be up and running for definitive agreements to be signed.

The next question is why project financing would be available for a novel battery technology with no QCP in place. There are two answers to this: 1) I have established this is a proven technology – see the 24M Kyocera contract below. 2) They have over $500m cash on their balance sheet which is both theoretically and according to IR sufficient to proceed to FID (Final Investment Decision) without project financing, which is expected to follow 6-9 months after FID.

In their Q1 presentation they already indicated they had already ordered some long lead-time items and that they were progressing to FID this year. Immediately following FID I believe they will be in a position to execute the Mpac option for at least the Gigafactory 1 part of 1&2 and so this may well occur on schedule in Q4.

FREYR Story

My focus will always be the numbers over the story, but it is important to understand what FREYR are about to try to identify some of the risks.

The first aspect is that their initial focus is on Energy Storage Systems (ESS). This could be for grid-scale storage to balance renewables primarily wind, or systems installed by individual consumers for balancing their own solar or peak electricity requirements. There is currently plenty of demand from ESS due to current energy market conditions but this being priced out by EV demand. The 24M technology is well proven for this application.

The second thing is FREYR’s location. Ironically cell manufacture is a very energy intensive process, including the assembly. 24M’s process reduces this, but a dry room is still required and FREYR’s location next to surplus renewable electricity supply is key to competitiveness. Furthermore there is some local access to raw materials and good shipping access.

The third positive is the Norwegian jurisdiction. This is ideal with their government consistently committed to renewables, inside EFTA with access to much EU grant funding, but without some of the political downsides of being in the EU.

So FREYR have a lot of natural advantages that make me think they are more likely to succeed than the average pre-production battery manufacturer. My main concern is that the second two advantages are so significant that they may be better off with a more proven, easier to finance battery technology than 24M.

24M

Much less is known about 24M, however we do know that Mpac are a “preferred supplier to 24M end-customers” and that Mpac have been working with them since 2018 (page 12 of MPac’s FY 2022 presentation). This opens up the possibility with contracts with other of 24M’s customers.

Using 24M’s press releases I will assess each announced customer in turn both as a test of the 24M technology versus competitors and as a possible customer of Mpac.

NEC Energy Solutions

This contract was signed in October 2015, after 24M “emerged from stealth mode” with their semisolid battery, and was to directly provide cells for assembly into storage systems.

In took a year for the initial shipment of cells to be made, with the next step being: “We look forward to working with the NEC ES team through this testing and validation phase as we also prepare the next delivery milestone in early 2017.”. That was the last they said about 24M, and in September 2019 NEC ES ordered 200MWh of “liquid metal” cells from Ambri.

Researching the Ambri technology it certainly seems impressive. But they pointedly fail to directly disclose their current or planned manufacturing capacity in GWh/year. The reason becomes clear with the disclosure that they plan to produce 200k cells a year by 2024 from their biggest plant and that a cell stores 1kWh. So that’s a mere 200MWh/year, smaller than FREYR’s CQP and a twentieth of their planned Gigafactory. Part of the issue appears to be that they have swapped lithium as the “exotic” element with antimony. While lithium is expensive there is at least a well developed supply chain whereas this needs to be developed for antimony with bulk US production not even scheduled until 2026. Another is that the current chemistry only seems to have been settled upon in 2021 whereas 24M’s main chemistry is mature with alternatives also demonstrated.

NEC announced in June 2020 they were closing down the company with speculation that battery commoditisation and thin margins due to heavy competition was the underlying cause and covid the final straw. The company was acquired by LG Energy Solution in September 2021 after the business environment improved.

Kyocera

In December 2018 Kyocera participated in a fundraising by 24M of $21.8. In June 2019 they began pilot production to validate use in ESS. In January 2020 they launched the Enerezza residential ESS in Japan and announced plans for mass production that Autumn. In May 2021 Kyocera provided further funding to 24M. I have confirmed Enerezza is currently widely available in Japan and the storage market in general is well established including with Kyocera as a supplier. There is a nice video with some explanation of the technology in this March 2022 video.

Clearly Kyocera do not use Mpac for cell casting and unit cell assembly. It is possible they have another partner, in which case Mpac have competition from a proven supplier, or they may do this in house, in which case equipment may not be available to third parties.

After research I am confident that Kyocera are not using any other technology for ESS, having previously resold Nichicon systems before making their own from the same Samsung SDI batteries and finally switching to 24M.

FREYR

I will include this for completeness.

In December 2020 FREYR became the first known independent licensee of 24M technology. In contrast NEC were to be supplied completed cells and Kyocera are a shareholder. In July 2021 they announced a contract with Mpac for cell casting and unit cell assembly machinery.

Although Kyocera mass produce cells, FREYR is potentially the first “Giga-scale” manufacturer. Initial production is expected in 2023 H2, now likely in Q4 at the earliest.

FREYR are open to using alternative technologies to 24M in future, leveraging their supply chain and cheap power, but they have no plans in place in the near term.

Fujifilm

Involved in the May 2021 funding round. If they plan to use the technology themselves rather than just collecting license fees then following the Kyocera example, production would take a further two years and is unlikely to involve Mpac. They do not appear to be involved with any other battery technology.

Lucas TVS

This is an Indian company originally formed as a JV between Lucas and TVS Group of India. Despite their origins, Lucas TVS are a very successful and innovative company producing highly-regarded products.

In September 2021 they signed a 24M license with plans to build one or more Giga-scale plants, with the first then expected to begin production in H2 2023.

Even 24M batteries are very energy intensive to manufacture and India’s low renewable energy production levels, no commitment to meaningfully improve and plans to double down on coal to reduce power outages is in sharp contrast to FREYR’s contracted 100% renewable 100% available supply. This combined with the Indian government’s recent political moves away from the West mean that neither supply by Mpac nor exports to compete in developed countries look likely. Indeed, current plans are to supply the Indian market only. Lucas TVS do not appear to be involved with any other battery technology.

Koch (KSP)

In January 2021 Koch Strategic Platforms (KSP) invested in FREYR.

In October 2021 they formed a JV with FREYR which made an investment in 24M in return for limited exclusivity for 24M’s technology in the US. Site selection will be in 2022 and targeted first production is by 2025. Initial and medium term production scale is likely to be similar to FREYR itself.

Mpac seems strongly positioned to win a manufacturing contract here:

• They already supply Kock’s JV partner.
• By the time they are likely to be choosing cell casting and unit cell assembly machinery, FREYR’s CQP should have proven Mpac’s machines. Indeed, they may be partnering with FREYR specifically so that their set up may be duplicated to avoid the need for their own CQP.
• Mpac have a presence in the US in the form of the Switchback subsidiary which should help satisfy any OO-ES-AY sourcing requirement.
• There is no clearly viable competitor to Mpac, though see Bosch under VW below.

Koch have made a number of investments in the battery space, some of which are synergistic with the use of 24M’s approach. One clear exception is an investment in Blue Current which is for a competing technology, but this is at a far smaller scale with only a megawatt-scale plant planned at this stage. The biggest risk I can see is that their scatter-gun approach to capital allocation means a lower probability that any one investment will succeed and an increased chance Koch implode.

Volkswagen Group

In December VW announced a strategic partnership with 24M to “manufacture next-generation lithium-ion EV batteries”. They then go on to say the focus is to “develop production technology for SemiSolid battery cells”. This involves an investment in 24M and formation of a wholly-owned subsidiary.

While this is further evidence of automotive interest in 24M’s technology which FREYR have been citing, this looks more like a potential threat to Mpac as it seems they are not happy with existing or soon to be realised 24M cell production technology. This contrasts with Tesla whose Gigafactory 1 defers cell manufacture to Panasonic using an utterly conventional process.

The January statement confirming the deal can be found here. A few days later they signed an MoU with Bosch to “explore the establishment of a European battery equipment solution provider.”, pointing out that “Bosch has excellent know-how in factory automation”. It seems unlikely that Bosch would let Mpac have such an important part of this, at least not without buying them. VW also said they planned to build six Gigafactories by 2030.

Regarding other technologies, in 2018 VW invested $100m in QuantumScape (NYSE:QS), a solid-state battery manufacturer. In April 2019 the company repeated that solid-state batteries are the future for EV, and forecast they would be available between 2025 and 2030. Further funding was provided in September 2020 when QS came to the market. QuantumScape’s pilot plant is now expected to begin production in 2023.

In the short term VW source from LG Chem, SK Innovation (SKI) and CATL depending on the end-market.

ARPA-E

Again in January, funding was announced from a US government department to develop batteries for aviation. The 24M process already works with multiple different chemistries and the US want them to look using a lithium metal anode which potentially has higher energy density.

I have ignored other grant funding, but this one is special in that it requires 24M to “develop and scale up a commercial, modular, pilot line”. This implies that the earlier research grant announced in September 2016 led to a potentially viable product. As a preferred supplier it is possible they may engage Mpac to help with production, although it is unclear whether the same equipment would deal with a metal anode and the minimum required scale is very small. I have found some further information here and a video here.

Inevitably the US government would prefer local suppliers to 24M, but Mpac do have always had some US presence and this increased significantly when the purchased Switchback.

Conclusion

This is going to be an unconventionally long conclusion! The following follows the order of the the risks identified in my previous article.

• 24M SemiSolid technology has been commercially sold by Kyocera since January 2020 apparently in increasing volumes. That means systems have been in use by end users for two and a half years at this stage. These are domestic systems made by a well known company in a developed market known for high safety standards so you can be fairly sure they are not catching fire or failing regularly. Assuming 24M are diligent then lab-made samples made from the same materials have likely been under continuous test by 24M from about March 2019, so well over 3 years now. Accelerated aging and stress tests have doubtless been done. But this is still a new technology and risks remain for products with expected lifetimes of 10-20 years.
• Mass production has been demonstrated by Kyocera, at least since late 2020, albeit not on a GWh/year scale. However this only proves that Kyocera have found it cheaper to continue running a plant where initial costs have been sunk, not that it is economic to build more, although announcements may have been made in Japanese that I have missed.
• NEC (now LG) Energy Solutions found that, or at least were persuaded that Ambri’s “liquid metal” high-temperature technology was superior for ESS, but that has made very slow commercial progress, appears to be well behind 24M, and also is not applicable outside of commercial ESS applications. VW seem to be tiring of the wait for solid-state and have hedged their bets with investment in 24M for EVs. Koch are investing in everything, but the only JV for manufacture is with 24M. Note that I have only researched alternatives amongst known 24M licensees.
• Bill Gates has thrown his weight behind some competitors, but seems aware that flooding the wrong technology with capital would be counterproductive. Conventional Lithium Ion and LFP batteries continue to improve and may present the greatest threat in the short term. But with the VW deal 24M does seem to be gaining momentum. Again, I have not done exhaustive research as this is a massive area.
• FREYR’s CQP is currently a financial quarter behind schedule and their ability to execute remains unproven and will effectively remain so until both this is complete AND they receive project funding. This is a high risk, but Mpac are not exposed to it in the short term as they are likely to receive the next contract ahead of QCP completion and project financing.
• On further consideration it seems almost certain that Mpac made a prototype machine in the three years they were having discussions with 24M, reducing the risks of failure. It is now clear that FREYR have placed their bets on Mpac as sole supplier and if they fail at the CQP then FREYR will be sufficiently likely to fail or be so late that whether they persist without Mpac is of little significance. If Mpac indeed succeeds at the CQP stage then they are highly likely to be used for purchasing decisions made subsequently – there would then be no other proven independent suppliers and switching would be risky.
• Revenue numbers for the initial contract are a matter of public record and can be extrapolated for the Gigafactories.
• Gross profits for the CQP are reflected in reported and forecast numbers and can be extrapolated to Gigafactories, although there is a margin risk around increasing input costs.
• The same applies to net profits.
• Payment so far has been 80% up-front, avoiding issues with the cost of capital, and also reducing the risk around the development of faster machines.

Summary

At current prices Mpac currently includes a cheap or zero cost option of revenues and gross profits doubling for a year, with a strong chance in that event of at least one further similarly sized contract from FREYR. I would then also estimate a 50:50 chance of one or more contracts from the Koch joint venture and an outside chance of a contract from or acquisition by Bosch.

The situation should be resolved within 4-9 months and I currently estimate a 50% chance of some further success, but interim developments that could be announced at any time may increase the probability of that significantly. Most likely either the share price will rise to reflect the value of this option ahead of time, but alternatively investors may remain ignorant to the magnitude of the upside allowing further cheap purchases closer to the decision point / as further news is announced. As has been seen previously in situations like this, excitable investors could push Mpac’s share price up to almost any level with or without due cause.

The best place to contact me and to discuss Mpac is Small Caps Live, although feel free to use the tools provided here also.