Future battery technology implications for xEV uptake

Future battery technology implications for xEV uptake
23 September 2020

In this post, we will look at the development of battery technology, chemistries & engineering and their potential implications on the future uptake of Battery Electric Vehicles and Plug-In Hybrid Electric Vehicles.

Public Health England says air pollution is the biggest environmental threat to health in the UK, however, the recent pandemic has led to a significant decrease in noise level along and concentration of air pollutants as vehicles were left unused for several months. Now that measures are easing, the future purchasers of passenger cars may lean towards an electric vehicle to keep this more pleasant environment. However, electric vehicles still account for 1-2 % of all vehicles on a global scale, and there are still major hurdles to overcome such as range capabilities, fast charging capabilities, cost reduction and safety concerns.

Which battery chemistry do OEMs currently use?

Current high voltage battery packs have mainly two different battery chemistries: nickel cobalt aluminium (NCA) and nickel manganese cobalt (NMC). NCA chemistries are typically used in all of the vehicles sold by Tesla while NMC is typically used by the other OEMs such as Mercedes, General Motors, VW, BMW etc

Decreasing cobalt content while improving energy density

Both chemistries contain cobalt which is typically associated with human rights and child labour issues as three quarters of all cobalt is extracted from the Democratic Republic of Congo. Moreover, cobalt is an expensive metal which has undergone significant price fluctuations over the last few years.

Lithium-ion cell manufacturers have been working hard to decrease the amount of cobalt in their respective cathode whilst ensuring continuous cell performance improvement. For example, Tesla has reduced the cobalt content of its batteries to 2.8 %, down from 5 % in their first cell chemistry whilst other cell manufacturers are now offering NMC 811 which contains 10 % of cobalt, down from 20 %.

To further increase this shift, and to increase the energy density of the battery packs, thus the range, several alternative chemistries are being developed which could displace current technologies. These technologies are namely, solid-state battery, high voltage chemistry, lithium iron phosphate, lithium-sulphur chemistry and sodium-ion chemistry.

High voltage cathodes chemistry

In high voltage chemistries, the current cathode (NCA or NMC) is replaced by another material that has a higher potential, thus increasing the cell energy density. There are several materials that could be used as a substitute, as shown in the table below:
lithium-ion battery cathode materials
Table 1 Potential high voltage cathode material for lithium-ion batteries (source)

Increasing the cell voltage would mean less cells are required to meet the pack voltage, leading to more cells being able to be packaged within the same volume, thus increasing the overall pack energy density. Though these materials show great promise, they still have significant issues that must be resolved to become a viable option. The most critical issue is that of long term stability, and it is yet to be comparable to the conventional lithium-ion battery material. Firstly, some of these materials suffer from manganese dissolution, especially at high operating temperatures, so there is a need to develop a robust strategy to mitigate this failure mode. And, secondly, a stable electrolyte has yet to be found to work in such an oxidising environment (source). These materials are still in the R&D phase at universities, so they are not going to disrupt the current lithium-ion technology before the next 10+ years.

Solid-state chemistry

Alternatively, solid-state chemistry has received a lot of interest in the last few years. In this chemistry, the volatile organic solvents are replaced by a solid compound whilst the anode and cathode remain the same as that in conventional cells. The solid electrolyte is typically an ionic liquid or polymer which still allows lithium-ion to migrate through it. The solid-state battery brings in welcoming advantages such as the mitigation of thermal runaway, which is one of the main concerns surrounding lithium-ion technology and its application in automotive. Another benefit is that the cell energy density increases. However, the major drawbacks are that the power density of the cell drops significantly and that, as for the high voltage cathode cells, so far, there are only a few start-up companies developing the technology. They have managed to prototype a small pouch cell system, but none of them are remotely close to mass manufacture. For example, Toyota just announced that limited solid-state manufacturing capability will be available by 2025, whilst Panasonic revealed 2 years ago that solid-state chemistries may not be available before 2028.

Lithium-sulphur chemistry

Lithium-sulphur technology is also very attractive as sulphur is a by-product of the petrol industry which means that it is a cheap commodity and that the environmental issues and its handling are known. Moreover, the cost of technology implementation is low, as it is possible to re-use what is currently being used to produce lithium-ion cells. In addition, the technology makes use of non-flammable electrolyte, thus mitigating against thermal runaway. Finally, the cell could be operating at higher temperatures reducing the requirements on the cooling system and potentially reducing the system complexity and cost.

However, despite all these advantages, lithium-sulphur suffers a number of drawbacks. Firstly, during the discharge process, sulfur is converted into several polysulfides which have high intrinsic resistance, thus limiting the discharge capability. Secondly, due to solid phase formation at the end of the discharge, a greater overpotential is needed to dissolve them through a redox shuttle mechanism, creating a low system efficiency. Finally, lithium-sulphur chemistry does not have a lifecycle that is comparable to conventional lithium-ion cells.
cell discharge curveFigure 1 Typical discharge curve for lithium-sulphur chemistry showing the different chemical reactions occurring as the material is discharged (adapted from source)

Sodium-ion chemistry

Sodium-ion chemistry is also getting some attention as it replaces lithium with a more abundant element, namely, sodium. In addition, it comes with similar benefit to that of lithium-sulphur in that this technology can be dropped in on the current lithium-ion manufacturing production lines without any large capital expenditure. Finally, it solves an important question regarding the safe transport of batteries as they can be transported completely discharged.

However, transporting the cells at 0 V also means that the cells, modules or packs would have to be charged on the production line, adding capital expenditure and potentially increasing production times. Additionally, sodium-ion cells offer about half of the energy density of current lithium-ion cells.

Lithium iron phosphate (LFP) chemistry

With the recent announcement that Tesla had chosen CATL as an additional battery supplier in China, it looks like LFP chemistry will enter the BEV and PHEV scene. LFP as a cathode material has some advantages over the other chemistries. For example, it has a higher breakdown temperature (~310°C) compared to NMC (~180-250 °C), and its energy release rate is the lowest of all the cathode materials. However, the cathode material is only part of the safety equation, and there is a need to include several additional factors to make a “safe” cell. Battery safety should include manufacturing quality, material purity, environmental factors, material coating etc. So, there is a need to take a holistic approach to safety and assure both operational and manufacturing quality.

LFP does not use cobalt in its make-up and consequently, has considerable advantages, however, the cell cost in $/kWh is higher than that of NMC or NCA. Additionally, once packaged into a battery pack, the state of charge estimation is not easily extracted through voltage measurement due to its voltage plateau, and the inaccuracy in the state of charge estimation can be quite large. Moreover, the cell nominal voltage is ~3.2 V compared to 3.6 V for NMC. Consequently, LFP cells have an inherently lower cell energy density to that of NMC cells. However, the voltage plateau can be very useful when selecting other e-powertrain components because the voltage change during the charge, and discharge will be less than that of NMC.

Fast-charging and the anode chemistry

Improving fast-charging capability is linked to 2 main factors: firstly, overall pack voltage and second the anode chemistry. The pack voltage dictates how much current flows through the battery pack during the charge. The higher the pack voltage, the lower the current, and therefore the lower the cooling requirements, which reduces the complexity and cost of the system.

To understand the anode chemistry, we need to understand the failure modes that we are trying to mitigate during fast charging.
The failure mechanisms are:
a) creation of lithium dendrite through lithium plating 
b) electrolyte decomposition due to high localised temperature 
c) electrolyte decomposition due to low or high localised potential at the anode and cathode, respectively

Electrode coating is not homogeneous so, in some areas, it is possible to observe a high localised current density. This would result in the local environment to reach a higher temperature and a lower or higher potential than the surrounding environment. Depending on the temperature reached, the electrolyte may start decomposing, and the cell would start swelling. Similarly, as the potential of the electrode increases or decreases for the cathode and anode, respectively, the decomposition potential of the electrolyte may be reached, and the cell would start to swell. Consequently, it is of importance to monitor coating thickness during the manufacturing process post electrode calendering.

Lithium dendrite is caused by the slow or blocked diffusion path in the graphite on the anode side. There are 3 potential cases to consider: a) charge at low temperature. b) high charging C-rate c) high SoC charge.

At low temperature, the diffusion of species is greatly reduced. If one was to assume that the charge rate is not changed between 25 °C and sub-zero temperatures, it is likely that there will be an accumulation of lithium at the electrolyte/electrode interface, caused by the blocked diffusion path. This blocked diffusion path is caused by the slow diffusion rate of lithium (intercalation and diffusivity) within the material. Similarly, at high charge rate, there would be an accumulation of species at the electrolyte/electrode interface caused by blocked diffusion path as the lithium would not have time to diffuse through the material before the next lithium arrives. Finally, at high SoC, the graphite is close to being full, so there will be an accumulation of lithium at the electrolyte/electrode interface caused by blocked diffusion paths or slow diffusion rate within the graphite material. These failure mechanisms are summarised in Figure 2. These diagrams explain why charging capabilities are typically expressed up to 80 % SoC by OEMs. lithium dendrite formation
Figure 2 Parameters affecting the formation of lithium dendrites on graphite anodes (source)

Improving anode chemistry could have 2 outcomes. Firstly, it could improve the energy density by replacing current graphite with higher capacity material such as a silicon/graphite anode mix. However, this has proven difficult, Increasing the amount of silicon yield to the disintegration of the binder material is difficult due to the large expansion that silicon undergoes during lithiation (up to 400 %). Consequently, only a small amount of silicon can be added to the anode mix to ensure the current binder technology does not fail, thus limiting the improvements that can be achieved.

However, adding a small amount of silicon still allows for significant anode capacity to be gained and to improve fast charging capability, some of that capacity may be given up by increasing the porosity of the anode material. Increasing porosity means that there is more contact surface area between the anode material and the electrolyte which yield to more sites available for lithiation, which in turn leads to faster-charging capabilities.

Alternatively, graphite can be replaced by lithium titanate, but this would result in a lower cell voltage, thus a lower energy density and a significant cost increase. Consequently, this chemistry is not used in the automotive sector.

Battery cost reduction

Over the last 10 years, battery pack cost, $/kWh, has fallen dramatically from more than 1000 $/kWh in 2010 to ~200 $/kWh mark in 2020. It has been suggested that once the $/kWh reaches below 100, electric vehicles will be affordable for the masses. At the above rate, it has been estimated that this threshold should be achieved by 2025. However, these will greatly depend on 2 things: firstly, battery pack engineering and cell cost.

In a typical battery electric vehicle, half of the total weight is that of the cells meaning that the other half is only there to protect the cells from damage (e.g. from a crash) and to connect the cells together and to safely manage them (electrically and thermally). Consequently, there is a need to optimise the battery pack design and decrease the amount of non-critical components within a battery pack. One solution could be a cell to pack approach in which the notion of the module is removed and the cells are dropped into the battery pack enclosure directly. This approach would yield to fewer parts and bring additional packaging efficiency gains, thus improving the overall energy density of battery packs whilst decreasing their cost.

A cell to pack approach would also bring in new challenges. For example, only cylindrical or prismatic cells may be integrated into such a design due to their rigid outer cell casing. That said, the prismatic cell requires a constant external compression to ensure electrical and ageing performance, so the higher the number of cells in a row, the higher the compression force that is required. Additionally, having all these cells within a battery pack means that there will be a need to re-think the approach to interconnect the cells using current welding technology. It is much easier to weld a module of 300 mm in width compared to one that is 1300 mm in width. Only 2 issues are cited here, although more could be covered. Both engineering and manufacturing ingenuity will be required to get the ‘cell to pack’ approach to become feasible for the mass market.

Cell cost is difficult to determine though there has been some information released by General Motors when the Bolt was launched that the 60 Ah cell cost 145 $/kWh. However, as the electric vehicle market increases, the utilisation of the cell production line will increase along with the number of these. Therefore, a further cost saving due to economy of scale will result in lower cell cost. Additionally, cheaper raw materials highlighted at the beginning of the post could also play a role in the reduced costs that batteries will experience in the future.

OEMs working together to create the future battery platform

One last thought that I’d like to leave with you is that potentially, it would be good to have more OEMs working together to develop a common EV platform, and packs and modules. I understand that this may sound like a crazy idea, especially from an OEM perspective but, do consider this from a customer and market penetration perspective. Firstly, everyone would get access to the state-of-the-art battery chemistry, and the cost of the vehicle will be much lower than that of today due to the economy of scale. There would be an improved utilisation of battery pack production resulting from OEMs collaborating on developing battery production plants together, rather than developing their respective production line in different parts of the world and not necessarily working at full capacity. Progress in this area has begun as VW is sharing its MEB platform with Ford, but these companies did not collaborate on the development. VW did the engineering work and is now licensing it to Ford for a better return on investment. Ford will need to work around what VW developed.

A question that will remain here is that of performance and competitive advantage, i.e. how could brand set themselves apart? Well, the answer to that question could lie in customer service levels, options provided in-vehicle, design and brand reputation/appeal. This approach is already taken for the internal combustion engine, whereby several engines find their way into different car brands, but the vehicles are still priced differently. So, if it is possible to do this on engines, why wouldn’t it be possible for batteries? Similarly, this approach could be used for e-motors, inverters, DC/DC and on-board chargers.

Conclusion

Though new battery chemistries are being developed for the automotive market, none of them are currently ready to displace current lithium-ion technologies. That said, in such a dynamic market, disruption is always possible.

As we covered earlier in this blog, improvement of pack engineering should be the main focus as this could yield to immediate improvement of pack energy density. This is something that the current EV leader, Tesla, is doing very well. For example, they have broken through the large OEM typical requirements in terms of maximum module voltage by having module voltages greater than 60 V and they also integrated the charger and the DC/DC converter within the battery enclosure. I am persuaded that future battery packs will follow their lead as it improves the offering without much engineering work.

I want to leave you with one final point, which is that I truly believe that if there were better coordinated efforts from governments and OEMs, the electric vehicle uptake would sky rocket! In such a scenario, OEMs could:

  • Freely share their customers’ experience with electric vehicles
  • Collect vehicle driving data
  • Work together to develop common battery packs and improve test standards

Government could provide reasonable subsidies and develop legislation and standards around these. This would yield to batteries having to pass tests based around real world data. This would help to create vehicles which are affordable and vehicles which are suitable for most of our daily travels (with the exceptional long mileage trip being performed in an internal combustion engine vehicle). Obviously, these should not stop at the batteries but at the complete infrastructure covering the charging infrastructure ,supply chain of raw material and recycling. These are topics we will look to explore in other blogs!