The economics of batteries
By Apoorv Shaligram
In the previous blog, I made a case for using the cost of ownership as a metric for measuring carbon emissions associated with any product or technology. Now, we are used to seeing costs as they apply to us (which are often impacted by discounting or subsidies. If we want to use it as a measure of carbon emissions, we have to consider the total cost (as in sum of all costs paid by consumer, company, govt. or whoever). Towards the end of the blog, we considered a comparison between EVs and ICEVs where we saw that EVs win in terms of cost over life, but are falling short significantly when it comes to upfront costs. Given that energy source and powertrains apart, EVs and ICEVs are mostly the same, this shortcoming in terms of costs is attributed primarily to the cost of batteries. In this blog, let us look at why EV batteries are so expensive and what it will take to bring these prices down.
Why are EV batteries so expensive?
One of the reasons why EV batteries are expensive is because of the rarity of the elements that go into it. Lithium ion batteries (LIBs) that are the first choice for EV applications are made up of Lithium, Nickel, Cobalt, Manganese, Graphite forming the active components along with Copper, Aluminum and Stainless steel covering the inactive components. However, expensive elements are not the only cause behind the cost of LIBs. LIBs are expensive because of the level of complexity of technology that goes into it. The active materials (cathodes and anodes) that go into making batteries have very tight tolerances on purity and specifications to deliver the performance expected of these batteries. These materials also have complex manufacturing processes; e.g., The NMC811 cathode chemistry that is used in the high energy density batteries today requires a high temperature calcination and heat treatment manufacturing cycle in pure oxygen flow conditions. The “cheaper”, safer and longer lasting LFP batteries use a LiFePO4 cathode material. The LiFePO4 powders need to be manufactured in an inert atmosphere and hence needs highly purified nitrogen flow in the calcination/heat treatment process. The graphite anodes used in these batteries should be quite cheap given that graphite is a widely occurring natural mineral, right? Dead wrong!!! For LIB application, the graphite should be highly purified, with surface treatment, particle shape modification and a tight size tolerance. Moreover, natural graphite does not give great performance for high end applications and hence, a large amount of graphite used in LIBs is artificial graphite formed from petroleum coke through an expensive and energy intensive high temperature process. Among the inactive materials, the electrolyte requires very high purity, especially with regards to moisture content which needs to be below 10 parts-per-million!! The copper and aluminum foils are around 8–20 micrometers thick. The separators are membranes made from polyethylene and polypropylene. At least, these should be cheap? No. Wrong again. These membranes need to be of the exact same thickness which is in the same range as the metal foils, but in addition should be uniformly porous, and the pore size has to be very very uniform, while remaining in the nanometer size scale. And these are just the materials we are talking about.
The actual cell assembly process comes next. A highly complex manufacturing process, with extreme quality control (QC) requirement that is also quite an energy guzzler (Refer previous blogs here and here to understand the LIB cell design and manufacturing process). Apart from the large energy cost of manufacturing that arises from a LIB’s “allergy” to moisture, the complexity, accuracy and sheer size of the machinery and equipment needed for manufacturing means that there is a HUGE capital investment in setting up a LIB cell manufacturing unit. While this reduces with scale, here is an example to understand its enormity: a Gigafactory with a manufacturing capacity of 10GWh/yr needs a capital investment of US$1B!! This is the number even after considering such a large scale! About 15-20% of the revenue of a battery cell factory goes into servicing the depreciation of manufacturing machinery and equipment, which is huge even when compared to any other manufacturing business. When seen in light of the fact that the input materials are themselves so expensive and the amount of energy cost of manufacturing, 15–20% begins to sound ridiculous. To put it in perspective, if a lead acid battery were to be manufactured in the same way as LIBs are, they would probably cost 2–4x of what they cost today. So, for people who feel that LIBs are expensive due to Lithium alone and switching from LIBs to NIBs (Sodium-ion batteries) will magically cut down costs, sorry, but it won’t as the cell design and the process is still the same and the problems associated with moisture remain as well.
In the entire value chain of LIBs, it is perhaps only the battery pack assembly process from cells, that fits through the window of conventional economic sense. Everything else is more a case of the there-is-no-alternative (TINA) factor as of today.
What will it take to bring it down?
Well, if it were just one major contributor to the cost, it would have been easy to target (If wishes were horses…). Nonetheless, let’s look at what are the possible ways of bringing the prices down.
- Chemistry: The chemistry selection comes first. We need to look for a chemistry combination that offers best value for money. Until a few years back, the race was on for the highest energy density possible as lighter batteries meant better energy efficiency for the vehicles. However, it is now becoming fairly clear that the added cost of making lighter batteries is not offset by the energy savings over life. In that sense, selecting a chemistry like LFP over NMC family makes a lot of sense. LFP is primarily Iron (Fe) based and as such the base mineral cost remains low. The process cost needs to be optimized.
- Design of application: An alternative approach would be one where a battery chemistry is picked that allows us to achieve application requirement with a smaller size battery. e.g., For most EV 4Ws, the design range lies between 200–500 km, while a single non-stop drive range hardly crosses 100–150km. So, if a battery tech allows a really quick charge without affecting safety or life of the battery, the design range of EVs can safely go down, thus reducing the cost of batteries per vehicle!
- Process: This is the most difficult one to solve for. The energy requirement for manufacturing comes mainly from the dehumidification of the battery cell manufacturing facility. While people have reduced this by splitting the facility into different zones which have different humidity level requirements, the only way to really reduce this energy level is by reducing the “floor space to manufacturing capacity” ratio. The second culprit in the energy consumption is the slurry coating process. When electrodes are coated with slurry, the slurry has to be dried off by heating. Tesla is looking to solve this problem (via its acquisition of Maxwell and holding their patents) by a process of dry mixing and extrusion. This solves for two things simultaneously: reducing floor space requirement, and energy requirement for coating and drying. It also partially solves the Capex issue by replacing the huge slurry mixing and coating machinery with smaller dry mixing and extrusion machinery.
- Cell form factor: Tesla is also looking to an optimum cell form factor. A large form factor cell makes life a lot easier from a manufacturing productivity point-of-view, but creates quite a headache from a QC and performance perspective. So, we want the QC and performance of a smaller cell (or perhaps an even lesser QC sensitive process), but the manufacturing productivity of a large form factor cell. Is there a better approach than what Tesla is working on? Yes. But it needs out-of-the-box thinking.
Note: The slurry coating technique for making electrodes is more than 30 years old and was earlier used in manufacturing of audio cassettes, video cassettes, camera films etc.. It is a question in itself as to why that technique is being used for manufacturing cells even today.
So, what else can we do? Quite a bit, actually. Though, if I go any further on this question, I would be would end up disclosing proprietary knowhow.😅 So, I guess, we stop here for today… But, one thing for sure. Someday fairly soon, we will see cell technology that reduces the cost of EVs quite dramatically! 🤟🏾🤟🏾
