Good day, and welcome to QuantumScape's discussion of the achievement of their final milestone for 2021. John Saager, QuantumScape's Head of Investor Relations, you may begin your conference.
Thanks, Operator. Good afternoon, and thank you to everyone for joining our discussion today. To supplement today's discussion, please go to our IR website at ir.quantumscape.com to view our press release and our Twitter page at QuantumScape to supplement the discussion. Before we begin, I want to call your attention to the safe harbor provision for forward-looking statements that is posted on our website and as part of our quarterly update. Forward-looking statements generally relate to future events or future financial performance or operating performance. Our expectations and beliefs regarding these matters may not materialize. Actual results and financial periods are subject to risks and uncertainties that could cause actual results to differ materially from those projected.
The safe harbor provision identifies risk factors that may cause actual results to differ materially from the content of our forward-looking statements for the reasons that we cite in our Form 10-K, recent 10-Q, and other SEC filings available on sec.gov, including uncertainties posed by the difficulty in predicting future outcomes. Except as otherwise required by applicable law, the company disclaims any duty to update any forward-looking statements. Joining us today will be QuantumScape's Co-Founder, CEO, and Chairman Jagdeep Singh, and our CFO, Kevin Hettrich. Jagdeep will provide a brief overview of the importance of achieving the goals we laid out at the start of the year, and then we'll kick it over to the analysts for questions. If time allows, we'll ask some questions we've received into our IR inbox. With that, I'd like to turn the call over to Jagdeep.
Welcome, everyone, and thanks for taking the time to join us today. Almost exactly a year ago, we showed the world the first solid-state lithium metal anode battery that was capable of working under what we refer to as uncompromised test conditions. As manufactured, the cells used an anode-less lithium metal architecture designed to deliver energy density of approximately 1,000 Wh/L , about 15% higher than the cells used in today's leading EV batteries, which are closer to about 715 . What was even more significant about that announcement, and this is what we mean by uncompromised test conditions designed to meet automotive specifications, was that it delivered 800 cycles to over 80% energy retention while running at fast one-hour rates of charge and discharge at 30 degrees Celsius, with less than five atmospheres of pressure, in a commercially relevant area of 70 by 85 mm.
This was a breakthrough announcement because, to our knowledge, no other group had ever shown equivalent performance data under these conditions for lithium metal anode cells. It gave us confidence that we had a fundamental technology with the potential to transform the automotive sector. However, this data was from a single layer cell, and we said at the time that a key remaining task for us to fulfill this potential was the need to scale up the battery by stacking up multiples of these single layers to achieve our commercial targets. To be candid, at the time, there was considerable skepticism regarding our ability to stack these layers up. Some were concerned that by stacking layers, we would somehow adversely impact the performance of the cells.
Others were concerned that we would not be able to stack layers mechanically, or the volume expansion that occurs in each cycle from the plating and stripping of lithium on the anode would prevent us from stacking. It therefore gives us great pride that this week, we were able to report that we have not only made 10 layer cells, but successfully achieved or exceeded those same performance parameters we reported on a year ago. More specifically, we showed data from a 10 layer cell that is, in what is to our knowledge another industry first, delivered 800 cycles with greater than 80% energy retention and did so at charge-discharge rates better than 1C 1C, a one-hour charge and discharge at 25 degrees Celsius, with 100% depth of discharge and modest pressure of 3.4 atmospheres.
The important point to note is that all of these parameters were met simultaneously in the same cell. This is key because a common technique used in the battery industry is to compromise on one of these parameters and then run other tests that compromise on one of the other parameters, never showing a single cell that meets all the requirements simultaneously as a real-world EV must. There has been a lot of activity in the battery industry lately. Some of the other approaches we have seen are attempting to use a liquid electrolyte with a lithium metal anode. The data reported from these approaches convinces us that they still suffer from the same types of problems liquids have had with lithium metal since the 1970s: dendrite formation and impedance growth.
To prevent the formation of catastrophic dendrites in a cell flooded with a combustible liquid electrolyte, these approaches have been limited to low rates of charge, for example, C/ 5 or five-hour charge, but only been able to show a short cycle life, insufficient to meet the requirements of the automotive sector. Others have shown cells based on sulfide electrolytes and published data, which has also not shown the ability to run at high rates of power at room temperature with long cycle life, all essential requirements for an EV battery. Thus, judging from the data we have seen, we believe such approaches have not yet shown they are capable of meeting basic automotive requirements and will need to overcome fundamental challenges at the materials level if they hope to be commercially viable. Some players are even trying to make larger cells with these liquid or sulfide chemistries.
However, we believe trying to make larger cells with a chemistry that doesn't meet the basic requirements is not a successful strategy. The fundamental chemistry issues will not fix themselves when larger cells are assembled. With this week's results, we have now met all of the milestones we laid out for ourselves at the beginning of the year. Our announcement this week gives us more confidence and belief than ever that our technology is a clear leader in the next-generation battery space. Going forward, we will continue to build on this lead by further improving our technology, manufacturing capability, and customer engagement. This includes increasing layer counts, introducing more automation, and refining our manufacturing process, building on the improvements to quality, consistency, and throughput that we have made over the past year and continue to make.
In 2022, we will focus on delivering prototype cells to our automotive OEM customers, as well as tooling up our pre-pilot QS-0 facility to produce cells for test vehicles in 2023. These goals require consistent execution from our team, and 2021 has shown that execution is one of our most important strengths. We're immensely proud of what we've accomplished in the last year, and we're looking forward to sharing more details in the coming months. With that, I'd be happy to open it up for questions. Operator?
Thank you, Yusuf. As a reminder to all participants, if you have a question, you can ask it by pressing star one on your touch-tone telephone. Again, that's star one on your touch-tone telephone. Please stand by. We'll compile the Q&A faster. Your first question is from the line of Rod Lache with Wolfe Research. Your line is open.
Hi, everybody. Thanks for hosting this call. I just have a couple of things. One is, I know that your target is to achieve the 1,000 Wh/L , but can you maybe just talk about where you are now? What's the corresponding energy density? How close are you to that target?
Hey, Rod. Thanks for the question. The energy density of the final cell is a function of two things. One is the energy density of the active stack, and two is the weight and volume of the inactive components of the cell, the packaging and other components that don't contribute to energy. The active stack, of course, is determined by the number of layers that you have in the cell. Each unit cell is one layer of active stack. If you have a single layer cell like we showed last year, the weight and volume of the packaging dominates the cell, and the energy density overall is not going to be at the targets that we have targeted. As we make four and now 10 layer cells, the overall energy density of the cell gets better.
When we get to our target of multi-dozen layer cells, that's when we'll get to the 1,000 Wh/L . We haven't really disclosed the energy densities of these intermediate cells, Rod, because they're not really sort of interesting relative to the targets that we have. That will really be achieved when we get to the final form factor, which is going to be, as I said, multiple dozens of layers in the package. The way to think about it is that if we take the architecture that we have right now for the single layer cells and increase the number of layer counts, that architecture, taking into account all of the packaging and other elements that go into a cell, we believe gets us to on the order of 1,000 Wh/L .
Okay. Just to be clear, it's basically just the packaging at the weight of these inactive materials now. That is just basically simple math. If you were to extrapolate from what you have right now, you would get to 1,000 Wh/L ?
That's largely true. There are other things that we plan on doing that are engineering improvements to the stack itself. The biggest element that is involved in going from where we are today to the 1,000 Wh/L is stacking up those layers into a multi-layer stack and decreasing the fraction of inactive materials to active materials in the cell.
Okay. Lastly, can you just give us some color on what your future milestones are from here? What we should be looking for beyond the 10 layer cell? Obviously, as you're going on to multi-dozen layer cells, is there some kind of timeline that you can just talk about as you progress towards your A sample?
Yeah. The two biggest things that are going to happen next year, A, is in fact this multi-dozen layer cell that we're targeting. That's going to be that won't be a big bang kind of a thing where we go from 10 layers to multiple dozens. We'll have incremental layer counts in various releases between now and the end of next year. Once we have those multiple dozens of layers, that will be what we call the customer prototype that we'll deliver to the customers. In parallel with that, we are going to be building out the QS-0 production facility. There are a lot of tools that we ordered. These are those high volume, high throughput tools that have long lead times. Those tools are arriving on various schedules over the course of next year.
Our plan is to try to get all those tools in the facility, qualified, turned up, commissioned, and ready to produce cells in 2023. In 2023, we start to ramp up actual production in QS-0. Obviously, it won't be from zero to full production overnight. Any manufacturing facility involves a ramp. The key requirement that we've set for ourselves is to be able to produce cells in that pre-pilot facility that address both the quantity of cells that are needed for our automotive customers, as well as sort of debug and test out the whole production concept with the facility. When you turn up a production facility, as we worked with our QS-1, Troy mentioned with Volkswagen Group, you really don't want to make any mistakes that you can avoid making. Our goal is to make all our mistakes, if you will, in this QS-0 pre-pilot line facility.
By the time we get to the real production facility, we have a much smoother and more streamlined process.
Okay. Great. Thank you.
Absolutely.
Once again, to all participants, if you have a question, please press star one now. Again, that's star one on your telephone keypad. Your next question is from the line of George Janarita with Barrett. Your line is open.
Hey, guys. Congratulations. Just to focus on next year's milestones, obviously, customer prototype sampling is the big one. You've mentioned multiple times so far in Q4 that that requires dozens of layers of cells. Can you kind of help us understand the time that takes? It took you months, you know, to get from four and then to 8 and 10. Is this something that gets easier as you add more layers? Should we expect something maybe around the middle of the year in terms of prototype samples?
We haven't tried to provide more precision on the dates other than to say our goal is to do this next year. I think relative to the difficulty involved, we do believe that the most difficult part of the process is to go from zero to one layer because at one layer, that's when that decision goes from not having the chemistry that works to having the chemistry that works. Going from one to four layers was a big step because the question becomes, "Okay, can you stack these things?" If you do that, does the volume expansion, do the mechanical interactions, do the electrical interactions between the layers in any way adversely impact performance? Obviously, we were happy to see earlier this year that that was not the case.
The question became, "Okay, can we continue that trend, add more layers, and see similarly good behavior?" I think what we're happy about now is that as we expected, because there are no chemical interactions between layers, stacking these layers up does not in any way adversely impact the capacity retention of these cells. We were able to produce cells that delivered the cell that we showed, in fact, this week that delivered 800 cycles to well over 80% capacity retention at really high rates of charge over 1C, 1C, so faster than one-hour charge and discharge. Going forward, we need to keep doing more of that. What does that involve? First and foremost, it requires making more material. We have to produce more separator films so we can make bigger cells.
Obviously, this 10 layer cell means that each cell that we make now uses 10 times as much material as the cells that we were making last year, which were single layer cells. If we get to multiple dozens of layers, that's another order of magnitude increase in the capacity of our engineering lines. Because we don't yet have the pre-pilot production line up, we're making all these cells on the engineering lines, which are primarily designed to conduct experiments and improve the overall technology and manufacturing process. We're now using them to actually produce cells in more volume. We're increasing the throughput of those lines as well. That was getting new tools, hiring new people to be able to run those lines. Those are all things that we're working on now.
Outside of getting more throughput, more capacity, there obviously is a mechanical engineering task involved to stack more layers up. One of the things that we have talked about that is a part of our manufacturing process is the notion of automated stackers that can stack these cells, these layers up more quickly, more precisely, more efficiently than we can with R&D tools. Those stackers are a key part of our production process. I think that, you know, you have to be, if you've been around long enough, you have to be a little bit humble in the face of the unknown. You can't just declare that there's zero risk going forward. I think we feel like the risk that's behind us really does demonstrate that what we have demonstrated already is that it's possible to stack these layers up.
It's possible to make a chemistry that meets what we believe are the core requirements of the automotive sector in terms of weight and cycle life and temperature and so on. It's possible to stack those layers up and not have issues with the expansion and contraction as the layers plate and strip lithium. We're going to keep doing that the rest of 2022. Our belief is we're going to be able to execute. Of course, we'll have to see if we're really doing in 2022 what we did in 2021, which is to meet all our goals. Thank you. Absolutely.
Your next question is from the line of Jose Asumendi with JP Morgan. Your line is open.
Hi, Jagdeep. It's Jose. Congrats on the.
Hey, Jose. How are you?
On the announcement. Good. Thank you. A couple of questions, please. I was wondering if you could talk a little bit more around what have you managed to change in the process or solve from the technology point of view to get to this 8 - 10 layer layering cell. What have you changed on the technology side? Did you get any reaction from Volkswagen in this announcement or that other OEM you seem to be working with? Did you get any reaction from the OEM with regards to this announcement? Thank you.
Yeah. In terms of what's changed, I think the two things that I would point to, one is we've been able to make more separators. We have increased the throughput of our process. We're going to continue trying to do that going forward. Secondly, we've improved the quality of the film. We alluded to this on the July earnings call, the Q2 earnings call, where we said that we've made some significant improvements to the quality and consistency of these films. That plays a critical role in actually being able to successfully make multi-layer cells because, as you might imagine, if you have 10 layer cells, every one of those 10 layers now has to work. If you have films that, you know, there's even one bad film, that's going to not allow you to make a 10 layer cell.
Quality is very important, and it's a key focus, has been a key focus of ours this year, will continue to be a key focus going forward. Relative to Volkswagen Group and others, I think the key point is that for all the customers that we have, and obviously, there's Volkswagen Group who also announced the second and top 10 automotive OEM, as you know, a few weeks ago. For all those customers, there's an ongoing set of deliverables that we plan to give them in terms of higher layer counts and more functional cells between now and when we have QS-0 pre-pilot production cells available. There are no other sort of events that occur relative to things like the $100 million investment that Volkswagen Group made back in Q1 based on testing cells.
This would be just us delivering them cells so they can test them, validate them, continue down the path of joint development and with the goal of actually trying to get these cells into real production vehicles.
Thank you. Thank you.
Sure.
As I'm showing no further questions over the phone lines, I'll hand the conference over to Mr. John Saager for additional questions from the IR inbox.
Yeah. One thing, if I can add, in terms of things that need to be done. We published data showing, for example, Rod, to your question earlier, if you're still there on the call, that the cathode loading that we've historically shown is on the order of 3 milliamp hours per gram. We'll probably increase that cathode loading over time as well to further increase the energy density for the same reason, that the cathode loading is active material. The more active material you have in the cell, the better the energy density of the cell going forward. The layer count increase is probably the single biggest factor relative to improving the energy density of the cell. There are some small improvements in other aspects of the stack, and the cathode thickness is one example. Let me turn it over to John.
All right. Thanks, Jagdeep. Thanks, everyone, for taking the time to hear our story. We're very proud of the team for hitting this last key milestone for 2021, and we look forward to continuing to execute into 2022 and beyond. For a more technical discussion of our third-party test results that we released on our Q3 call, we have an event tomorrow at 10:00 A.M. You can find details for that event on our Twitter page and our LinkedIn pages. Thanks, everyone.
This concludes today's conference call. Thank you for joining. You may now disconnect.