Solid-state batteries - The science, potential and challenges

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
Hi. I’m Dr. Billy Wu and in this video we’ll be talking about solid-state batteries; covering the science behind them, their potential but also the challenges that still need to be overcome. Firstly, why are we interested in solid-state batteries? In order to understand this we first need to look at the conventional lithium-ion battery. These devices power our everyday electronics, electric vehicles and provide large scale energy storage for the integration of renewables. However one of the major drawbacks is their modest energy density, which impacts the range and energy efficiency of your electric vehicle, with this currently at approximately 250 Wh/kg. Also, in addition to this, there are safety concerns around their use. So, with this in mind, lets now look at solid-state batteries, or more specifically solid-state lithium-metal batteries. Firstly, these have improved safety since we use different materials in their construction. Their theoretical energy densities are generally higher, with values in excess of 400 Wh/kg, which can unlock new applications such as electrified flight And they can be charged faster However, whilst these theoretical advantages are very attractive, there are lots of practical barriers that still need to be overcome. Now before we take a deep dive into how solid-state batteries work, lets take a step back and look at what happens during thermal runaway in traditional lithium-ion batteries since safety is one of our key drivers. In a conventional lithium-ion battery we have an anode which is traditionally made from graphite and a cathode made of a transition metal oxide, which is separated by a polymer separator. All of these components are porous which allows a liquid electrolyte to soak all the regions which allows lithium-ions to move between the 2 electrodes. The graphite anode is, however, thermodynamically unstable when in contact with the liquid electrolyte, which causes a solid electrolyte interphase, or SEI, to form. This is a protective layer which limits further reaction between the anode and the electrolyte. If the battery heats up to over roughly 70°C, the SEI layer begins to decompose leading the electrolyte to react with new anode surfaces. This is an exothermic reaction which causes more heat to be generated. If the heat is not removed and goes over around 130°C, this can cause the polymer separator to melt which causes a short-circuit between the anode and cathode, leading to more heating due to the large current flow. Above about 200°C the liquid electrolyte decomposes, and above 300°C the cathode decomposes, leading to further heat release. The cathode in particular is important, as its decomposition also leads to oxygen being given off which is extremely flammable. Here, you’ll note that a key component of the thermal runaway reaction is the flammable liquid electrolyte, which is a major safety issue. So motivated by the safety challenges posed by liquid electrolytes, as well as the desire for higher energy density batteries; research and development in solid-state batteries has been increasing in recently years, but what is a solid-state lithium-metal battery? Well, as we said before, traditional lithium-ion batteries have 2 porous electrodes which work via the insertion, or intercalation, of lithium into their atomic structures. Here the organic liquid electrolyte allows the lithium-ions to move between the electrodes, with electrons flowing around an external circuit to provide useful electrical work. In the case of a solid-state battery, the liquid electrolyte is replaced with a solid electrolyte which ideally is inflammable and safe. In doing this, it also allows for the graphite anode to be replaced by lithium-metal which has about 10 times the specific capacity, giving the potential to increase cell level energy density. However, a major problem here is that when cycling the battery, the lithium-metal is stripped and deposited continually, which over time can lead to void formation, lithium dendrites or cracks in the solid electrolyte which can compromise the performance. Another important consideration arises when we consider how a lithium-metal battery works. During discharge, lithium metal is stripped from the anode to become a lithium-ion and electron. This lithium-ion then moves to the cathode to intercalate and the electron goes round the external circuit to extract useful electrical work. When the battery is charged, much of the lithium has been stripped meaning that the thickness of the battery is smaller. This reaction is then reversed when charging to plate the lithium back onto the anode. However, as mentioned, doing this repeatedly can lead to flaws in the lithium and solid-electrolyte. Thus, one of the key takeaways is that there can be significant volume change in a lithium metal battery, which often requires additional compression to address for good lifetime and the volume of the cell is state-of-charge dependent, which can affect its stated volumetric energy density. Next, lets have a look at the different solid-state battery architectures and their pros and cons. Starting with the lithium-ion battery, this technology is now mature with supply chains scaling up and a relatively easy manufacturing process, having benefitted from decades worth of research and development. However a major drawback of the technology is its relatively modest energy density and the safety concerns associated with the organic liquid electrolyte. Next we have the all solid-state battery where the liquid is replaced by a solid. This has the key advantage of being safer. However, the drawbacks of this is that the manufacturing is more difficult due to the need to sinter the solid material or challenges with handling the material. Furthermore, because we now have a solid instead of a liquid, the interface between the active material and electrolyte is more difficult to maintain since the solid does not flow as readily. Finally, in this configuration with a traditional graphite electrode the energy density is actually lower. This final point is important, as simply replacing the liquid electrolyte with a solid actually reduces the energy density as you are replacing the liquid with a solid which can have a density which is almost 5 times higher. This can, however, be addressed by replacing the traditional graphite with a higher specific capacity anode material such as silicon. However, in this video we’ll focus mostly on lithium-metal based all-solid-state batteries, whereby the graphite anode is replaced with a lithium metal foil. This has the advantage of being potentially safer than liquid electrolyte batteries and affords a higher energy density cell because of the use of lithium. However, this is still difficult to make because of the challenges with consolidating the electrolyte with the cathode active material which also leads to a higher resistance and thus a lower power battery. Also, the reversibility of the lithium-metal anode still remains a problem. To get around this, hybrid solid-state batteries have been proposed, whereby a solid electrolyte is still used to separate the anode and cathode, but a liquid or gel based catholyte is used on the cathode. This retains the high energy density from the lithium-metal, improves the cathode-electrolyte interface giving a lower resistance and also makes the battery easier to make than the all-solid-state battery. However, since a organic liquid, or gel, electrolyte is used, there are still some safety concerns as well as the fact that excess lithium-metal is used which is intrinsically reactive. Finally, we’ll cover the hybrid anode-free cell, where no lithium-metal foil is used in the cell construction. This works because when we first assemble the battery, the cathode already has lithium in it, which can be plated onto the copper current collector on the first charge. The advantage of this approach is again the good interface between the cathode and electrolyte but now that the lithium-metal foil has been removed this affords one of the highest energy density configurations and also removes one element of the manufacturing process, making them easier to make than other variants. However, the drawback is that there are still some safety issues with the catholyte and the reversibility issues associated with the lithium plating and stripping is even higher, since there isn’t any excess lithium used. Other variations on these designs do exist but in order to keep this video concise, I’ve left these out. Now regardless of the configuration, common to all solid-state batteries is the fact that they replace some degree of the liquid electrolyte with a solid. Here, there are various types under development but I’ll focus on the three most commercially mature ones. Firstly, we have sulfide based electrolytes which generally have excellent performance characteristics in terms of lithium conductivity. Their mechanical properties are reasonable but because they are somewhat plastic, it makes them relatively easy to coat and process. However, the disadvantage is that they have relatively low oxidation stability, poor compatibility with traditional cathode materials and generate hydrogen sulfide when in contact which water, which is a poisonous, corrosive and flammable gas, giving them poor safety characteristics. Next we have oxides, which have a lower conductivity but are generally more stable leading to excellent safety characteristics. However, because these oxide materials often require sintering to get good conductivity this makes them difficult to process, with the ceramic films generally being quite fragile and prone to fracture. Finally, we have polymer electrolytes, with polyethylene oxide, or PEO, being one of the most researched types. These organic polymers generally have a poor conductivity and thus have to be operated at higher temperatures, however their processability is excellent since they easily flow and are quite durable. However, these polymers have limited thermal stability and generally decompose at relatively low voltages, limiting the choice of high voltage cathodes. So, given all the potential benefits of the technology, why aren’t we all using solid-state batteries? To start off with, lets get a better understanding of the practical energy density of a traditional lithium-ion battery. Previously, we’ve already gone over the structure of a battery, with the main components being the anode, cathode, separator, electrolyte and metallic current collectors. However, when we look closer we can see that in both electrodes, we also have a conductive carbon additive which facilitates electron movement. And also, we have a polymer binder which allows all the components to stick together after its been coated. Now if we only consider the theoretical performance of the active materials you can see we have a battery with a gravimetric energy density of over 600 Wh/kg and almost 2,000 Wh/L which is extremely impressive. However, in reality we lose some of this theoretical performance to practical considerations such as the need for excess anode material for good lifetime, electrode balancing or first cycle losses. Next, we also need to consider the additional mass that the electrolyte introduces, which can be quite significant especially with certain types of solid-state electrolyte materials. Then, we also need to consider the additional mass of the current collectors, binders and separator. And finally, we also need to consider the housing material, tabs and gaskets which are needed for a practical cell. When all of these practical considerations are considered, you can see we’ve gone from an energy density of about 600 Wh/kg and 2,000 Wh/L to a more realistic value of ~260 Wh/kg and 630 Wh/L In terms of what components are important you can see from this pie chart that the cathode, anode and electrolyte materials are major contributors to the cell weight, and hence there’s been significant research in these areas. The energy density losses, however, don’t stop there; and when we look at a battery pack there are additional losses. Here, if we look at data from a teardown of a Volkswagen ID.3, the cells have an energy density of 273 Wh/kg and 685 Wh/L. When scaling to a module and pack, the additional weight of the inactive components such as the housing, cooling system and electrical components decreases the pack energy density to 173 Wh/kg and 284 Wh/L. Finally, because of the additional losses in a battery pack from the interconnection resistance and parasitic power consumption of other components, this further drops to a measured pack energy density of 162 Wh/kg and 266 Wh/L, which is clearly far from the theoretical potential of the materials, highlighting the opportunity for system optimisation. So, lets now return to the case of the all-solid-state lithium-metal battery example. Here, the graphite has been replaced with a metallic lithium-metal anode, the liquid-electrolyte replaced with a solid-electrolyte and often polymer binders in the cathode aren’t needed as these are bound together with the solid-electrolyte. If we only consider the theoretical performance and mass of the active materials, this gives an energy density of over 1,000 Wh/kg and 5,000 Wh/L. However if we consider that we often need excess lithium in the system for good lifetime and other practical constraints this decreases to just over 700 Wh/kg and 2,400 Wh/L. Next if we add the mass of the electrolyte, this can significantly decrease the energy density and is also a strong function of the electrolyte thickness, where a thinner layer is ideal but harder to make. Then if we include the mass of the current collectors and carbon black. As well as the mass of the casing material and tabs, we reach an energy density closer to 400 Wh/kg and just over 1,100 Wh/L. This of course is a welcome improvement on traditional lithium-ion batteries but far from their theoretical performance. When looking at a mass breakdown, we can see now the major factors are the cathode active material and the electrolyte, so careful selection of the cathode material will have a profound impact on the cell level energy density. So, if we now put these 2 analysis next to each other we can see that whilst both battery types lose theoretical performance when considering the constraints of a practical cell, the solid-state lithium-metal cell does have improvements over traditional lithium-ion batteries. In this specific example we looked at, this resulted in a 49% improvement in the gravimetric energy density and 80% improvement to the volumetric energy density. Acknowledging that these numbers will vary for different configurations and materials used. Of course energy density isn’t the only consideration and one of the major barrier holding back solid-state batteries is their degradation modes. Here there are many but I’ve just focused on a few notable ones. One of the main degradation modes is the fact that the solid-electrolyte does not perfectly block lithium filaments or dendrites from forming when charging. Here, we want a nice and dense plate on the anode but lithium can often grow through the electrolyte causing a short circuit if it reaches the cathode. When discharging the battery and dissolving the lithium, this can lead to interfacial delamination which causes spots on the anode to lose contact with the electrolyte. When cycled multiple times, we can cause surface flaws and voids in the lithium which is not ideal. Also dead lithium can form in the electrolyte which leads to capacity loss since its no longer electronically connected to the anode and not participating the electrochemical reactions. Another consideration is that lithium is a relatively soft metal and as such we can get mechanical creep of the metal which can cause issues such as short-circuits between layer and uneven anode loading. Asides from the lithium, the electrolyte also has various issues such as stability when put in contact with the anode and cathode, whereby a mismatch can lead to the decomposition of materials. At the cathode side, depending on the operating conditions this can also lead to gas generation which can cause the cell to inflate and delaminate layers. In certain solid-electrolyte materials, they are also very brittle and the stresses in the battery can cause fracture and cracking. This is also an issue in the cathode which often expands and contracts during cycling which can lead to cracking of the cathode and loss of active material. In the case of solid-state batteries where the electrolyte is not as flowable, this expansion and contraction can lead to void formation and loss of contact with the electronic and ionic conducting phases. Finally, the cathode can undergo various phase transformations which lead to a loss in performance but this is also true for lithium-ion batteries. Thus at the system level additional compression compared to traditional lithium-ion batteries is often added which increases system complexity and mass. Because of the losses on the anode side, excess lithium is often used to get a decent lifetime which adds additional mass and cost. And finally because of the issues with maintaining a good interface at the cathode, a liquid or gel catholyte is often used which can potentially increase safety issues. My final point before I conclude is just to highlight the rate of innovation and scale-up in the battery field. The Joint Centre for Energy Storage Research in the US have defined the battery technology readiness levels, or BTRLs, to highlight the various stages of a new technology. This first starts with BTRL-1 where there is a scientific breakthrough or new innovation. This is often followed by BTRL-2 in which the new class of material is synthesized which can take about 2 years. From here at BTRL-3 we want to prove the performance of this material in small coin cells which adds a few more years. Once, confident in the results, at BTRL-4 we then develop larger cells to prove the performance which are more representative of real cells, but might revisit the earlier synthesis and performance. Then, once we have this confidence at BTRL5-6 we then focus on how to scale-up the materials, manufacturing and performance at the pack and system level. All-in-all this can take upto 10 years to fully realise. Finally, if we are going to scale any new technology, we also need to consider the rest of the supply chain. This chart from Benchmark Mineral Intelligence highlights that setting up a new mine can take between 5-25 years, new chemical processing plant between 3-5 years, new cathode production 2-3 years, cell production 1-2 years and validating all the performance 5-8 years. These huge lead times and high capital requirements to develop out the supply chain, highlight the hurdles that need to be overcome to deploy a new technology at scale. So, to conclude, are solid-state batteries our next great hope for enabling our low-carbon future, or are they overly hyped. Well, we can see that if we replace the graphite in our batteries with lithium-metal this does enable much higher energy densities compared to traditional lithium-ion batteries. The replacement of the liquid electrolyte with a solid-state electrolyte can increase the safety by removing or reducing the amount of flammable components, and in doing this does make it easier to use a lithium-metal anode, however its important to remember that safety concerns still exist. Within the family of solid-state batteries, we have different variants from the all-solid-state battery, hybrid using some catholyte and anode-free, each with their pros and cons. Various different types of solid-electrolyte exist with the 3 most prominent being the oxides, sulfides and polymers. Sulfides generally have the best conductivity, but can form hydrogen sulfide when reacted with water and have some stability issues. Oxides are quite stable and safe, however are difficult to process and don’t have as good a conductivity, and polymers are the easiest to manufacture since they are flowable, however have poor conductivity. When looking at energy density numbers we need to be careful since theoretical numbers are always somewhat inflated and we need to consider the true system performance once the weight of all components are considered. In addition to energy density, we also need to address the various new degradation modes that lithium-metal batteries introduce. And finally innovations being reported now, will take time to scale-up with various challenges along the way. However, whilst this is quite challenging, strong progress is being made and I for one am hopeful for the potential of this technology. So, hopefully you found the information in this video helpful. I have other videos on other topics around batteries, fuel cells and materials on my channel so please do check them out or leave a comment if you have questions about batteries.
Info
Channel: Billy Wu
Views: 47,519
Rating: undefined out of 5
Keywords:
Id: NPaOJceBkJs
Channel Id: undefined
Length: 22min 13sec (1333 seconds)
Published: Mon Jan 09 2023
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.