The Promise of QuantumScape // First Principles Advantages
Welcome back everyone! I’m Jordan Giesige and this is The Limiting Factor. This is the third of three videos on QuantumScape – the good, the bad, and the ugly. The last two videos covered the bad and the ugly. I covered these first to clear the air so that we can appreciate the promise of QuantumScape’s technology. That is, today we’re gonna
look at the good, which is the first principles benefits of solid state lithium metal batteries. Before we begin, a special thanks to my Patreon supporters and YouTube members. This is the support that gives me the freedom to avoid chasing the algorithm and sponsors, and I hope will eventually allow me to do this full time. As always, the links for support are
in the description. Back to QuantumScape. Let’s start with why Solid State batteries are considered the holy grail of batteries. We first need to define what type of solid state battery we’re referring to, because there are a few different flavours. The specific type of solid state battery that we’ll discuss here is what’s called an anode-less solid state battery. What do these words mean? Solid state is straightforward. In a conventional
liquid electrolyte battery cell, the lithium ions travel back and forth between the cathode and anode through a liquid. In a solid state battery, the liquid electrolyte is replaced with a solid. This offers greater safety because liquid electrolytes are typically flammable. Next, what’s anode-less mean? A conventional lithium ion battery uses copper coated with graphite for the anode. The graphite is a stable host structure that’s used to store lithium at 1 lithium atom for each 6 carbon atoms. With an anode-less battery cell, the anode is just copper with no coating. When the battery charges, lithium plates directly
to the copper as lithium metal. This is the most energy dense way to store lithium metal at the anode because no bulky host structure is required. In other words, anode-less solid state batteries are actually one hell of an innovation because they replace two pieces of the core technology in a conventional lithium ion battery – the anode and the electrolyte. Doing just anode-less or just solid state are each, in their own right, quite a feat. Doing both at the same adds to
the difficulty but it’s also why some people call solid state batteries the holy grail of batteries: It maximises both energy density and safety at the same time, two requirements which are usually mutually exclusive of each other. QuantumScape claims to have mastered anode-less and solid state simultaneously, and they’re claiming they’ve achieved excellent performance in every test dimension at low cost. Whether they can deliver on that claim with a commercial product remains to be seen, but for this video let’s assume they can and explore the technology.
Let’s look at QuantumScape’s Specs compared to what we might expect from Tesla in 2024, when QuantumScape is expecting to enter into low volume production. QuantumScape is projecting a pack level energy density of 228 Watt hours per kilogram assuming a 353 mile range. With some back of the napkin calculations QuantumScape is looking at a cell level energy density of around 350 watt hours per kilogram when their cells go into low volume production in 2024. This is excellent compared to today’s liquid electrolyte battery cells, which are around 260 to 300 watt hours per kilogram. 350 watt hours per kilogram is similar to what I
expect Tesla to be producing in 2024, and I expect Tesla to be leading the market. It’s worth noting that at 350 watt hours per kilogram, QuantumScape will be just beginning to mine the energy density potential of their battery cells and over time their battery cells should exceed far beyond 400 watt hours per kilogram. I expect the energy density of Tesla’s liquid electrolyte cells to max out at around 400 watt hours per kilogram based on the potential energy density improvements I see today.
Anode-less solid state batteries are able to achieve higher energy densities because they don’t have material coating the anode and they have slightly higher voltage potential than graphite or silicon. The issue instead tends to be cost, which we’ll cover later. By 2030, if Tesla wants to keep an edge in energy density, they would need either a better cathode, anode-less with a liquid electrolyte, or their own version of solid-state. The question is, will Tesla need greater than 400 watt hours per kilogram? In my view, only if they create an electric VTOL. Otherwise, cost will be the key metric. Next, charge rate. QuantumScape expects their batteries to be able to charge from 0-80% in 15 minutes. This excellent compared to today’s liquid electrolyte battery cells. Once again, I’m expecting similar performance from Tesla’s battery cells in 2024, but his all depends on how much Silicon Tesla can fit in their battery cells without causing cycle life issues.
In terms of cycle life how does QuantumScape perform? For a Nickel based cathode chemistry, 800 cycles would be a minimum target because for a battery pack that provides 250 miles of range that would be 200,000 miles over the life of the battery pack. Tesla’s battery cells for Nickel chemistries are good for 1200 cycles or more, which provides 300,000 miles over the life of the battery pack and that will increase over time. If we extend the lines from QuantumScape’s test data, 1200 cycles is achievable. This looks promising but we really need real world cycle life data to prove it out. What all this means is that in terms of performance, QuantumScape expects to match the performance of high end conventional battery cells in the mid-2020’s. In 2030 and beyond,
the first principles advantages of solid state batteries should come into play and establish a firm advantage in energy density for QuantumScape. But, energy density, charging speed, and cycle life aren’t the only factors to consider. The number one priority for auto OEMs is cost. QuantumScape is indicating that in 2028, they’ll be producing 91 Gigawatt hours with six and a half billion in revenue. This means they’re expecting to sell their battery cells at $70 per kilowatt hour at the cell level. QuantumScape is also claiming a 30% gross margin,
which would put their cell production cost at $50 per kilowatt hour. Hold that thought. By 2028, at the cell level, Tesla may be able to achieve less than $30 per kilowatt hour with LFP and $35 per kilowatt hour for high nickel chemistries. That’s a finger in the wind estimate, but certainly doable. This is because Tesla’s vertically integrating most of the supply chain and won’t need to pay sales taxes, logistics fees, or a profit margin to third parties. As for the price of cells in the broader battery market, the average price in 2020 was $102 per kilowatt hour. Over the past 30 years, lithium ion cell prices have dropped by 19% for each doubling of production capacity. If battery cell production grows at 50%
per year that means a cost reduction of 7.5% per year, or $59 kWh in 2028. That is, it looks like QuantumScape plans to sell their battery cells for a premium compared to a conservative forecast for conventional lithium ion battery cells in 2028. If their battery cells do end up having an edge in charging speed or energy density, auto OEMs may be willing to pay that premium. But, in my view, there needs to be something else to sweeten the pot. QuantumScape will have to provide other benefits that liquid electrolyte battery cells can’t. Luckily, they appear to have at least 5:
First, if the data provided in QuantumScape’s presentation doesn’t contain any fish hooks, they should have better low temperature performance than current lithium ion batteries. This is because liquid electrolytes begin to freeze up at low temperatures, whereas a solid-electrolytes are already frozen. There are solid electrolytes that perform poorly at low temperatures, but QuantumScape is using an electrolyte that should exhibit good high and low temperature performance. The majority of vehicle owners don’t have a gripe about cold weather performance, but again, we’re talking niche benefits that a small group of people are willing to pay a premium for. And, as far as niches go, this’ll be a big one. Cold weather performance is one
of my most frequently asked questions. Second, high temperature performance. Liquid electrolytes tend to start degrading at high temperatures and become explosive. Solid electrolytes actually perform better at high temperatures, which softens up the solid materials in a solid state battery, allowing better contact between the solid layers, which increases charge and discharge rates. This thermophilic nature means QuantumScape’s batteries might last longer in hot climates. But, it’s dependent on how temperature tolerant
QuantumScape’s cathode gel is. The cathode gel isn’t solid state, and if it isn’t stable at higher temperatures, it may reduce the safety benefits and high temperature performance. I note that QuantumScape’s high temperature slide didn’t appear to include the cathode, and only tested the lithium anode against the electrolyte. That is, it doesn’t appear to be
a full battery cell, which would have given us an idea of the performance of the cathode gel. Third, with better high and low temperature tolerance, solid state batteries require less thermal management. This allows simplification and weight reductions at the pack level. Fourth, liquid electrolytes are flammable. This means when the battery gets too hot or when the battery shorts out, it can burst into flames. In a solid state battery,
the electrolyte isn’t flammable, which is why solid-state batteries are safer. Fifth, because solid state batteries are safer, they might allow for the use of less packaging and therefore weight. This, combined with reduced thermal management requirements means that at the pack level, solid state batteries should reduce costs and increase energy density. Now that we’ve looked at the major specs, cost, and fringe benefits, how do we interpret the market for QuantumScape’s battery cells? Here’s how I think it’ll play out. Tesla’s 4680 battery cells, along with battery cells from China, will dominate the 2020s due to their lower cost vs solid state. However, if QuantumScape can deliver on the benefits of solid-state, they should be able to find a niche that’ll give them a foothold in the market. From there, they can expand on that niche as volume drives down production costs and the
first principles manufacturing benefits of solid state begin to play out, eventually driving the cost of their batteries below the cost of liquid electrolyte batteries in the 2030s. In other words, the first principles manufacturing benefits are where I think QuantumScape’s great long term potential lies. What are those first principles manufacturing benefits? First, as I said earlier, anode-less solid state batteries have the potential for higher energy densities because they’re able to completely eliminate graphite and silicon coatings on the copper current collector of a battery cell. This also means that during manufacturing, the entire coating process for the anode can be eliminated, meaning the chunk of capital to buy those machines and the floor space to house them can be eliminated. Next, because no electrolyte filling is required, further money and floor space is saved. Instead,
the electrolyte is just a ceramic sheet. After the ceramic sheets are shuffled into the other layers of the battery cell, the battery cell is capped and it’s ready to go to formation cycling. Formation is the stage of battery cell production where the battery is charged and discharged over the course of several few hours to finish the production process. After formation, the cell is aged for around two weeks, giving the cell time to exhibit defects before final testing. This means
millions of battery cells are just sitting on the shelf, not generating cash flow. All up, formation and aging account for about 25% of the cost of a battery factory and take up a lot of floor space, as shown in this image. QuantumScape claims that they’re able to mostly eliminate this step. To understand why we need to understand formation. In liquid electrolyte battery cells with graphite and/or silicon at the anode, formation as the name indicates, forms a protective layer during the first charge and discharge cycles. This process sacrifices expensive lithium to form a protective
layer on the graphite anode material. With an anode-less battery cell, there’s no material on the anode, so formation can effectively be skipped. QuantumScape claims that overall, these first principles improvements to the manufacturing process will result in a cost reduction of 17% compared to conventional lithium ion battery manufacturing.
What I find most encouraging here is that QuantumScape is attacking different cost reduction opportunities Tesla. If QuantumScape can implement solid state technology and also implement manufacturing technology similar to Tesla’s, they could eventually reach a production cost lower than what Tesla indicated at Battery Day. As I’ve said in the past, Tesla doesn’t have a monopoly on those technologies. For example, there are other methods of creating a dry battery electrode that don’t use Tesla’s hot roller technique, and therefore wouldn’t infringe on Tesla’s patents. The counterpoint is of course that Tesla will continue to innovate, and Battery Day is just the beginning. In my view, Battery Day was a
window into Tesla’s plans until 2024 and we don’t know what Tesla has planned for the late 2020s. 2024 where the story begins for QuantumScape and I expect it to be many years after that before they a) implement some of the innovations we saw at battery day and b) start moving upstream of battery cell production into raw materials processing to drive down cost. With all that said, the important point here is that an anode-less technology and solid state technology like QuantumScape’s aren’t just an energy density improvement. They’re manufacturing improvements with first principles advantages that I expect will eventually be adopted by all major cell manufacturers in one form or another, including Tesla.
In the last video, I was hard on QuantumScape and with good reason. They have big challenges ahead. But, are there indications that QuantumScape might actually succeed? In short, yes. First, it sounds like QuantumScape is brute forcing their solid state project and have run over 2 million tests in 10 years. They didn’t provide any information on the number of cells those tests related to, but for comparison, Sila Nanotechnologies tested roughly 35,000 cells over about a decade to arrive at their chemistry. QuantumScape claims that they landed on an
effective solid state material in 2015 and for the past 5 years they’ve been figuring out how to make it low cost and scalable. That is, it’s one thing to find the right material, and it’s another to find a recipe for that material that uses cheap raw material inputs and a cheap manufacturing process. I’m guessing that in the last year or two they were able to get the numbers to work on paper, and now it’s time to test the recipe and manufacturing plan against reality. This is significant because one of the primary cost drivers in solid state batteries is the solid state electrolyte, which usually runs about $2000 dollars per kilogram. For a solid state battery to be profitable, the cost of the material per kilogram needs to be closer to $20 per kilogram. But not only does it need to be cheap, it needs to be thin. If the solid electrolyte is too thick, even if it’s cheap, it’ll blow out the cost targets.
QuantumScape claims that they’ve found a formula using 4 cheap bulk materials and can manufacture a film that’s between 10-20 microns, which sounds like a home run. We don’t know the cost of their raw material, but if it’s available in bulk that’s a good start. With regards to thickness, I would have been happy to see less than 30-40 microns, so 10-20 is very promising. Even more impressive, they’re claiming that their solid electrolyte is as cheap as the separator that’s currently used in liquid electrolyte batteries to separate the cathode and the anode. This means QuantumScape’s technology allows the liquid electrolyte and plastic separator of a conventional lithium ion battery to be replaced with a solid electrolyte for the same cost.
The next thing that usually makes solid-state batteries expensive is that they require expensive manufacturing techniques. QuantumScape solves this problem by using a solid electrolyte that comes in the form of a free standing ceramic sheet rather than a complex coating process. They also claim that their ceramic sheets are robust and flexible.
Typically, any kind of manual handling, let alone bending, will introduce cracks into a solid electrolyte, creating defects which cause the battery cell to short out. I’d be curious to know if the solid electrolyte separator in this image was still able to be used in a battery cell. This much bending would usually crack many solid electrolytes, causing the cell to fail. QuantumScape claims a cell pressure of 3.4 atmospheres. This means for the cell we see here, they had to put it under about 200 kilograms of pressure during testing. Solid state batteries
usually require much higher pressures, such as 20 versus 3.4 atmospheres. 3.4 atmospheres should be manageable for pack engineering. Furthermore, QuantumScape has claimed that the battery cell performs well without pressure and that it’s only automotive applications that require higher performance and therefore pressure. What about the expansion of the cell as the lithium layer plates and de-plates from the copper current collector as the battery cell charges and discharges? The expansion and contraction of each cell will likely only be a few millimetres, which again, should manageable for pack engineering. Overall, both the pressure and expansion can be handled with proper pack design. It may add weight to the pack, but I’d expect that would be less weight than the solid state battery cells themselves are saving through reduced thermal management and safety requirements. That’s not to say the pack engineering will be easy,
just that the expansion and pressure requirements aren’t a showstopper. There are also criticisms that QuantumScape’s battery isn’t pure solid state. While that’s true, the question is: Does it make a difference? We don’t know without knowing the details of the gel electrolyte that they’ll be using at the cathode. There are two risks that I see with using a gel.
First, gel electrolytes typically aren’t as robust to high temperatures as solid state, which may reduce cycle life at high temperature operation. Second, if the gel touches the anode, it’ll corrode the anode, which is an engineering challenge. If both of these problems have been solved, then QuantumScape’s Semi-Solid cathode may perform as well as a pure solid state battery. I’m guessing the gel won’t be as robust as solid state, but will be more robust than a standard liquid electrolyte. Let’s summarise the three part video series on
the good, the bad, and the ugly of QuantumScape. The Ugly: The Scorpion report was a self-defeating report that took positive information out of context and twisted it into negatives, misinterpreted graphs and scientific data, made suggestions that would have required a time machine, and admitted in their disclaimer that their informants were typically paid, possibly biased, and the information they provided may have been outdated. The Bad: QuantumScape, for its part, opened the door for the Scorpion Report by announcing a breakthrough while not having a market ready product, in a presentation that didn’t consider the audience, in an overheated investing market, and just after the Nikola and Trevor Milton saga. It wasn’t a good look to say the least and they made a rod for their own back. The Good: The promise of QuantumScape is a new type of chemistry that can fill market niches in the late 2020s. When it eventually grows up in the 2030s it could go from a niche product to the primary way in which batteries are made due to the first principles manufacturing advantages of anode-less and solid state technology. Although new battery factories will likely use some form
of these technologies, will solid state batteries replace existing liquid electrolyte battery lines? History says no. Battery cell production capacity from new technologies is typically built on top of the old. This is why we still have lead-acid batteries, alkaline batteries, and nickel metal hydride batteries. Once a battery technology meets
the needs of a use case, investing money to upgrade battery lines with improvements that customers don’t need or want is a misallocation of capital. Additionally, further developments and improvements to liquid based chemistries will mean those production lines will continue to be competitive with solid state production lines for the next ten years in terms of both energy density and cost. Tesla fans should be rooting for QuantumScape and every other solid state company. Why? First, solid-state technology isn’t threat to Tesla’s dominance because even under a best case scenario, solid-state won’t become relevant until the end of the decade. Second, what we saw at Battery Day was a preview of what we can expect up to 2024, and we should expect that Tesla will find even more ways to extend the runway for lithium ion batteries or develop their own version of solid state. However, solid-state is hard, and even if Tesla doesn’t come up with their own version, with the multi-trillion dollar market cap Tesla might have at the end of the decade, they could buy-out the company that has the best version of solid state. Overall, a future that has solid state as an
option is a much better future, so I hope that QuantumScape does deliver. In the meantime, there are multiple improvements to liquid electrolyte batteries and newer chemistries such as sodium ion and iron air. That is, solid state isn’t the only game in town, and we have a lot more to cover on the channel. In the next video, we’ll look at the competition between aluminum and steel body structures and how Tesla’s gigacasting blows the entire cost model out of the water and replaces it with a new reality. If you enjoyed this video, please consider supporting me on Patreon with the link at the end of the video. I am also active on Twitter. You can find the details in the description, and I look forward to hearing from you. A special thanks to Mihec Bojc, Mark Kett, Brian
Kinstler, Jim Dennis for your generous support of the channel, my YouTube members, and all the other patrons listed in the credits. I appreciate all your support, and thanks for tuning in.
2021-09-24 07:46
