Source : Benchmark Mineral Intelligence – www.BenchmarkMinerals.com – September 2017
The scramble to build new lithium ion battery capacity is underway. In 2020, the world will have at least 265GWh of additional manufacturing capability, an increase of 230% from today.
Confusion reigns, however, between what constitutes a lithium ion cell plant and one that produces battery packs for electric vehicles and stationary storage devices. In quantifying raw material demand, understanding this difference is vital. Here, Benchmark Mineral Intelligence outlines the production process to make lithium ion cells and battery packs as the world prepares for the rise of the battery megafactories.
The lithium ion battery megafactories are rising. In February 2014, Tesla announced its Gigafactory – a battery plant that could produce 35GWh of cells, and 50GWh of battery packs.
It was an outrageously audacious plan at the time and one that would increase lithium ion production by more than 50%. At that time, there were only two other battery megafactories in the pipeline: Panasonic’s Dalian, China plant; and Foxconn Technology Group’s idea to build a facility in Anhui, China.
Foxconn’s plans never left the drawing board and instead the Apple iPhone manufacturer recently invested $1.2bn into Contemporary Amperex Technology (CATL) – China’s biggest lithium ion battery producer for 1.9% of the business.
Panasonic on the other hand has since started production at its new 5GWh plant in Dalian at the beginning of 2017 and is part owner of Tesla’s Gigafactory which also began production earlier this year.
Since the early days – albeit just over three years ago – a global battery arms race has resulted in 17 battery megafactories under construction totaling 265GWh of capacity, according to Benchmark Mineral Intelligence data.
There is little doubt about the impact that this will have on the upstream mining industry, those companies that mine and process minerals to battery-grade, specialty chemicals. The impact on the downstream auto sector is also clear.
There is also no dispute that the electric vehicle (EV) has arrived and the world is now in the cusp of an energy storage revolution. The dawn of the semi mass market EV now means consumers will soon have a choice of mid-priced, pure battery powered vehicles owing to the launch of Tesla’s Model 3, Chevrolet and General Motors’ Bolt, and Nissan’s new 60kWh LEAF.
Confusion reigns however between the manufacturing of lithium ion battery cells and packs. In the mainstream media, there has been somewhat of a pack mentality to brand every plant that handles lithium ion cells, a new cell manufacturing facility.
This year has seen the addition of three new plants from SK Innovation in South Korea at 4GWh, Dynavolt in Fujian, China at 6GWh of nickel-manganese-cobalt capacity, and a new Tesla-inspired megafactory plan from Northvolt in Sweden.
Benchmark has also upgraded CATL from 50GWh to 100GWh as it expands to meet major contracts with some of China and Europe’s biggest auto manufacturers.
Momentum has been added to this build out in battery capacity with Volkswagen Group (VW) recently revealing it will need at least 200GWh worth of lithium ion batteries by 2025 each year.
The numbers are simply huge. In total by 2020 the industry can presently expect an additional 265GWh of new lithium ion battery cell capacity – the first step in the battery production process.
Considering the relative speed in which a new battery plant can be expanded – 6 to twelve months depending on equipment availability – expect this new megafactory capacity number to rise over the coming years.
With the rapid expansion of cell capacity, a clear distinction must be made between these plants that actually purchase the raw materials and make the cells, and those that assemble the EV packs.
Preparing raw materials
There are three mainstream lithium ion cell designs: cylindrical, pouch and prismatic. Cylindrical designs look most like a battery that the world is familiar with: a larger version of the AA sized batteries that would go into our TV remote controls.
Whereas, the pouch and prismatic batteries are different shapes to the common perception and used in portable electronics and vehicles.
The life of all three cell designs begins in the same way: electrode fabrication. This is where the raw materials are mixed with solvents before being coated onto an aluminium sheet (cathode) or a copper sheet (anode).
Using the example of an NMC chemistry, nickel-cobalt-aluminium oxide powders are lithiated with lithium carbonate and mixed with solvents to create a paste. This paste is then applied in a micrometre thin layer onto the aluminium sheet.
A 5GWh plant would require 4,000 tonnes of lithium carbonate and 8,300 tonnes of NMC cathode material. Four 1,000 gallon mixers would be needed for the cathode material and solvents.
For the anode, graphite – either coated natural spherical graphite, micronized synthetic graphite or a blend of both – is mixed with solvents and pasted onto a copper substrate.
Around 6,000 tonnes of graphite anode material would be needed for the cell, requiring two 1,000 gallon mixers. It is likely, however, that the plant would consume up to 20% more graphite than what just goes into the batteries as it is also used as lubricant throughout the manufacturing process.
Once the anode and cathode are mixed and pasted into their respective substrates, the calendaring process begins. This is a series of high pressure rollers that press the raw material onto the substrate and ensure uniform thickness.
In this process, the thinner the sheets are, the better the battery as more raw material can be fit into a lithium ion battery cell increasing its density.
The final step in the electrode preparation slitting – a process which cuts the electrode roll into the desired width of the cell. Once this is complete, the raw materials are ready to enter the battery formation stage which differs slightly for each cell design.
Most common today is the 18650 – a measurement of the radius and height of the cell. Major producers of 18650 are Panasonic Corp in Japan which produce the cells that are consumed by Tesla’s pack assemblers in Fremont, California.
Future vehicles from Tesla – starting with Model 3 this year – will use a new cylindrical design, the 2170. This is a taller and slightly wider version of its predecessor which has an increased volume of 46%.
There are many reasons Tesla decided to redesign the classic 18650. Speed of manufacturing and plant density are key.
For Tesla to produce 500,000 electric vehicles each year, the speed in which the Gigafactory can churn out lithium ion cells is a critical factor.
The new larger battery design means less individual cells need to be produced for each vehicle while the battery pack remains the same desired capacity.
“The machine that builds the machine,” was how Telsa CEO, Elon Musk described the Gigafactory at launch earlier this year. “[Its] not a hodgepodge of things where the machines are bought from a catalogue.
There’s almost nothing in a Model S that’s in any other car… the same approach has been taken for the [Gigafactory],” said Musk “The output is going to be volume times density times velocity. What is the density of useful to non-useful volume?
Its crazy low [between] two or three percent” “Then you say velocity. What is the reasonable expectation? Carmakers may make a car every 25 seconds, it sounds fast, but length of car plus buffer space is five metres– it’s taking 25 seconds to move five metres.. you’re not much faster than a tortoise at that point.”
“The density improvement may be as much as an order of magnitude of improvement [from] two to three percent to 20 of 30 percent [volumetrically]. With significantly less engineering effort, we can make dramatic improvements to the [Gigafactory],” he added. Tesla’s first principals approach has drawn some criticism from experts in the battery who are experts in building traditional battery plants. One expert explained to Benchmark some basic errors were made when Tesla was first planning the Gigafactory including having some heavy battery equipment on second and third floors of the building, something that it could not structurally support in the designs.
However, the battery industry has evolved step by step for the last 25 years – adding plant capacity when needed onto the original building.
For the first time, the potential in battery demand has forced the rise of megafactories that now warrant rethinking the way they are produced. The first step in lithium ion battery manufacturing is winding of the cathode and anodes into ‘jelly rolls’.
While winding a 2170 sized roll will take more time than an 18650, it is still one of the quickest parts of the process. Significant amounts of time are saved in placing these jelly rolls in aluminium casings which must be sealed. Even more time is saved in the pack manufacturing process where each cell is riveted and secured into the modules which make the pack.
The second major benefit of the 2170 is the weight in the vehicle. Some of the heavier components in a battery are the aluminium casing and other related parts. Considering there at least 7,104 of the 18650 cells in a Model S 85kWh model, using the new larger cells could reduce this to under 3,850 for each car.
Reducing the amount of material used across the board and the number of cells in a pack saves both cost and time. Recent interest from new start-ups such as Faraday Future and Lucid Motors – new players that are seeking to follow a similar platform to Tesla – has set a path for the 2170 to become the premier battery design for EVs.
As a result, both of Panasonic’s long-term competitors, LG Chem and Samsung SDI, are now developing their own 2170 designs to meet new demand.
Pouch and Prismatic
Before Tesla’s rise, pouch and prismatic were the foremost cells in the EV space used in pioneering cars such as Nissan’s LEAF. Development of a pouch design allowed for larger cells than cylinders and did not require rigid and heavier casings for individual cells.
The manufacturing process begins in a similar way to its cylindrical counterpart with a winding or ‘folding’ process of the cathode and anode into a desired shape. Instead of forming a cylindrical jelly roll, the manufacturing equipment produces a stack formation, much like folding a piece of paper into four.
This stack is then encased in a thin aluminium foil creating a pouch which can be used individually in tablets, laptops and mobile phones. The prismatic battery design is larger version of the cylindrical jelly roll process which is then placed into a hard aluminium casing.
Samsung SDI have been the industry’s champion for this lithium ion battery design and have had a significant marketing push for this to be the favoured cell for the EV industry.
In 2017, the battery major launched a range of prismatic designs and capacities to target the sector with 28Ah, 40Ah, 60Ah, 64Ah options launched using an NMC cathode chemistry and blended (natural and synthetic) graphite anode.
The plants that produce these cells are fundamentally different to assembly facilities that make the EV packs. For example, Samsung SDI recently opened a new ‘battery plant’ in Hungary that caused much confusion in the market place.
The facility based in Goed, 30km north of Budapest has the ability to make packs for 50,000 EVs. “Batteries are one of the most important parts supplied to global car makers. I expect the plant to contribute much to the growth of the European EV market,” Samsung SDI president, Jun Young-hyun said.
The plant however will be using cells from Samsung’s battery plants in Korea and China. It may be producing cells and will not be procuring the raw materials that go into the cells – this process will happen in Asia.
In the circumstances when a pack facility does not produce cells, it may purchase an electrode roll which it would then make into cells however it will still not be procuring the raw materials. This is the case with Automotive Energy Supply Corp (AESC) which purchases its electrode roll from NEC.
The pack area of the lithium ion supply chain is a rapidly developing link in the supply chain. Mastering the pack or the platform is a critical component of making a high quality EV.
It is where the skilled engineers really earn their credit by decisions on the optimum capacity of the pack, the number of cells that should be used, the size and shape of the cells, the shape of the pack, the cooling systems that are deployed and many more.
In a pure EV, the pack is essentially the vehicle’s engine and is where the serious intellectual property ground can be gained or lost. Do strategies of the lithium ion battery majors differ?
This has been a recurring question that Benchmark is asked and it is quite interesting to observe the three different strategies that have been deployed. Samsung SDI has been one of the first to significantly invest in pack assembly outside of Asia.
Tesla started life as a pack manufacturing company rather than an EV producer and has since integrated upstream into making the cells at the Gigafactory in conjunction with Panasonic.
For Tesla, significant reduction of costs through supply chain integration was a critical step in its mission to turn the world electric. For Panasonic, teaming up with Tesla meant the cell manufacturer was vested in both pack and vehicle production.
LG Chem, the third battery giant of the group, has focused on expanding its cell manufacturing capacity in four global hubs: South Korea, China, Poland and the US. Expanding its presence from the home of South Korea was a deliberate strategy.
The Korean lithium ion producer is presently building its newest EV megafactory in Wrocław, Poland, in the south west of the country close to the border with Germany and Czech Republic.
It will have an initial capacity of 4GWh and construction is now underway with a start-up expected by Q4 2017. “We will turn the Poland EV battery plant into a mecca of battery production for electric vehicles around the world,” UB Lee, President of LG Chem said at the ground-breaking ceremony.
Raw material flows
To understand where lithium, graphite, cobalt and nickel will be consumed, a clear distinction needs to be made between where the cells are manufactured and where the packs are assembled. Benchmark launched its lithium ion battery megafactory tracker to focus on where the hotspots of raw material demand will be located.
The biggest cell producer by 2020 is expected to be CATL with a 100GWh capacity. This would translate to an annual demand for lithium carbonate of 80,000 tonnes, cobalt chemical consumption at 23,000 tonnes, and its graphite anode needs at 120,000 tonnes.
While these numbers are huge in comparisons to where the industry is today, more than a doubling of battery raw material demand in 2016, it is quite conservative as to where the battery industry is heading.
VW have been more vocal about their plans to produce EVs in recent months. The most significant news from world’s number one auto manufacturer was the revelation of its expected lithium ion battery consumption by 2025.
The group says it will require a minimum 200GWh of lithium ion battery cells each year from 2025 onwards.
Considering the lithium ion battery industry was only 70GWh in 2016, it is a significant step change for the sector and the first time a traditional auto manufacturer has been so explicit.
Benchmark does not expect VW to be making its own battery cells anytime soon, but significant investments into cell manufacturing capacity must be on the horizon for the German group.
In raw material terms, 200GWh translates to 160,000 tonnes of lithium carbonate, 216,000 tonnes of graphite anode, and 63,000 tonnes of cobalt chemical. This is additional supply that the industry alone would not be able to fund without commitments from the world’s biggest end users like VW.
Pack of the future
Battery packs are becoming bigger and denser. When the Toyota Prius hybrid was first produced in 1997 it had a 1.3kWh nickel metal-hydride battery, while Nissan LEAF began its life in 2010 with a 24kWh pack.
Today for a mid-range EV, this would be deemed a very small battery pack with Model S at 85kWh, Model X at 100kWh and Lucid Motors’ and Faraday Futures new designs at over 130kWh.
The new standard capacity for mid-range vehicles is likely to be between 60-65kWh with both Tesla’s Model 3, Nissan’s third generation LEAF, and Chevrolet’s Volt all aiming for these sizes.
This would give a range of over 230 miles on a single charge. When considering electric buses, the packs are even bigger. Proterra’s new designs contain packs that range from 79kWh to 660kWh.
The most common way to improve the pack is to make the cells denser. However, adding more cells means adding more weight.
The reduction of passive materials – those other than the cathode, anode and separator – in a battery pack is of paramount importance. Tesla’s new, bigger 2170 design is a good example of this trend and its one that reduces the amount of aluminium casing needed in the pack.
The cooling system is also another area of significant weight. Today’s EVs are liquid cooled therefore much research is going into air cooled pack designs. There is also focus on improving how the vehicle draws energy from the batteries.
Therefore, software – the energy management system – has a vital role in making today’s lithium ion batteries better.
While chemists have received the credit for the improvements we have seen in lithium ion batteries over the last 10 years, it will be the pack engineers and software programmers that will own the next decade. Maybe having a pack mentality isn’t a bad thing after all.