A couple of weeks ago, French Environment Minister Nicolas Hulot announced that petrol and diesel car production would cease in France by 2040. Last week UK Environment Minister Michael Gove followed suit on behalf of the UK. Currently electric cars are dependent on Lithium Ion batteries for their on-board store of energy and this obviously raises questions about supplies of Lithium (Li), and as luck would have it a comprehensive report on this topic fell into my mail box last week. The report Raw material needs by the Li-ion battery industry by Dr P. Kauranen  is available online and in this post I want to provide a simple and partial summary of that report whilst adding information from other sources.
- In 2015 global Li production stood at 32,500 tonnes / annum and reserves were estimated to be 14 million tonnes giving a nominal reserves / production ratio (NYSE:ROP) of ~431 years at current production and consumption rates.
- Global consumption of Li is projected to grow 4 fold by 2025 cutting the ROP to 108 years and consumption is expected to continue to grow beyond that. However, as demand and price rises so will the activity of mining companies and it is to be expected that reserves figures will rise too, perhaps broadly in line with demand.
- Li-ion batteries currently account for ~35% of global lithium demand and in 2015/16 electric vehicles accounted for 64% of Li-ion battery demand, the remainder being portable electronic devices - cell phones and computers. Hence EV's currently account for 22% of global Li consumption.
- In 2015, China had 32.5% of the Li-ion battery manufacturing market, the USA had 7.3% and the EU 3.5%. By 2018 this will have changed to China 42%, USA 32% and the EU 1.4%.
- The massive jump in the USA performance is down to Tesla's Gigafactory, which is designed to build 35 GWh of Li-ion batteries per year (see inset image up top) and scheduled to commence production in 2017. (Note that once complete, Tesla's Gigafactory could account for 28% of global Li-ion battery production. It is perhaps this statistic, rather than electric car production, that underpins the companies very high rating.) The Li-ion battery ambition of the EU is totally at odds with member states' declared ambitions to go 100% EV by 2040.
- The main uses of Li-ion batteries in ~2016 are as follows:
- Cellphone 19.0%
- Tablet computer 7.3%
- Laptop computer 10.2%
- Battery electric vehicle 28.8%
- Plugin hybrid EV 5.3%
- eBuses China 29.3%
- Note the staggering figure for eBuses in China where it is reported that 94,000 vehicles were produced in 2015! Furthermore, stationary storage like the Tesla Powerwall has not yet got off the ground.
- Li prices have more than doubled since the end of 2015 but have been subdued this year. With global demand perhaps set to quadruple by 2025 we can expect a roller coaster ride as new supply and new demand play off each other.
Dr Kauranen's report dated May 2017 appears to be a comprehensive and professional analysis. The host institution is Strateginen Tutkimus that is the Academy of Finland. However, certain inconsistencies leave me feeling a little uneasy. For example, the thousand separator is sometimes a "," sometimes a "." and sometimes nothing at all, a sign of sloppy editing. The units used are tons which in English refers to an imperial ton but I assume that it is metric tonne that is used. This has left me having to cross check a lot of the numbers to find that there is a vast range in web-based articles. For example, Wikipedia documents global Lithium production as 600,000 tonnes per annum while Dr Kauranen's figure is 32.500 ton/a which I assume means 32,500 tonnes per annum. Cross checking with other sources shows that Dr Kauranen's figures reflect reality.
Li-ion batteries have enabled the portable electronics revolution to take place because for any given size (volume and / or mass) they are able to store a lot more energy than their peers. The table below from this source suggests that Lithium ion offers 4 to 5 times the energy density of Pb-acid batteries, the nearest practical competitor. This translates to a handy range of 150 miles for an electric car compared with 30 to 40 miles using Pb-acid. The initial cost may be 5 to 10 times higher, but Li ion offers many more advantages such as higher cycle rates, deeper discharge with constant performance and an operational range at higher ambient temperature conditions.
Figure 1 A comparison of the operational performance of Li-ion and Pb-acid batteries. Lines 2 and 4 are key. The increased energy density (Wh/kg, line 2) makes electric vehicles viable while the increased cost (line 4) is over ridden by the advantage of line 2, at least for so long as generous subsidies for Li-ion EVs exist.
At this point the other important fact to know is that the majority of Li ion batteries today use a Li - cobalt (Co) compound in the electrolyte and the availability of Co is therefore of equal importance. There are alternatives to using Co such as manganese (Mn) which offers poorer performance but technology may be forced down that route if Co availability (price) proves to be a limiting factor (see below).
James Stafford at Oilprice.com provided a usefull review a year ago that included this graph (Figure 2) that illustrates the huge advantage that Li batteries have over competing technologies.
Figure 2 The energy density of various Li-ion battery technologies compared with nearest peers - lead-acid and nickel cadmium. X-axis Watt-Hours per liter (volume) and Y-axis Watt-Hours per Kg (mass).
A Note on the Chemistry of Lithium
Lithium is an alkali metal (single positive charge) and will share chemical properties with sodium (Na) and potassium (K). This is why one of the major sources of Li is brines that occur naturally within porous rock. Li with atomic weight of 7 is the third lightest element after hydrogen (weight=1) and helium (weight = 4) which are both gases. Lithium is a metal which clearly places it in the "special" chemical category.
A Note on Ore Paragenesis and Global Mining
Li is mined either from rocks that contain the mineral spodumene (a Li pyroxene) or from brines that are enriched in Li. Figure 3 (from here) shows the global state of play.
Figure 3 Global Li production according to country and ore paragenesis: rock versus brine circa 2008.
We see that brines account for ~50% of global supply (~2008) and that Li brines occur mainly in S America. In China and the USA both brines and rock ores occur. Rock dominates everywhere else, expecially in major producer Australia.
Figure 4 Table 1.
Kauranen reports Li production of 32,500 tonnes / annum and reserves of 14 million tonnes providing a nominal reserves over production (ROP) figure of 431 years at today's consumption rate. There is clearly no immediate concern about Li resource availability. Of course consumption rates are expected to grow substantially and the ROP figure may fall rapidly as a result. However, the reserve and resource base will also grow and I will guess that the ROP figure will not have changed much in 50 years time.
An alternative source (Mohr et al.) estimates Li resources to lie between 19 and 55 million tonnes with the best estimate of 24 million tonnes. Kauranen's figures overlap this range. Mohr et al. estimate production of the order 26,000 tonnes per annum in ~2011, again broadly verifying Kauranen's figure.
One important observation from Figure 4 is the ROP figure for Co is 72 years using the reserve figure of 7.1 million tonnes. At face value, therefore, Co is in shorter supply than Li, but using the resource figure of 120 million tonnes the ROP rises to 1,200 years. Not much to worry about there.
Where there may be some justification for concern is on security of supply of Co. Li has diverse sources - Chile, Argentina, Bolivia, China and Australia - while Co supply is dominated by the Democratic Republic of Congo which has a long history of geopolitical unrest.
Figure 4 shows that batteries consume 35% of global Li and 42% of global Co production.
Finally, Figure 4 shows a recycling rate of 68% for Co but only <1% for Li. I don't understand why Li has such a low recycling rate and improving this will obviously have major impacts on future supply and price forecasts.
Where Li-ion Batteries Are Made
Figure 5 Table 2
It should come as no surprise to learn that China leads global Li-ion battery production with 32.4% of the global market. The USA currently has 7.3% and the EU 3.5% market share using figures in column 1 (Figure 5) -- i.e., fully commissioned plant.
The units are MWh which reflects the number of MWh of storage manufactured each year. To provide some context an entry level Tesla Model S has a 75kWh battery pack. Global production of 51,549 MWh per annum is sufficient to power 687,320 new Tesla Model S vehicles per year.
Column 5 (the Total) in Figure 5 is actually the sum of columns 1 to 3. If we assume that announced capacity will get built (the figure of 35,000 is Tesla's Gigafactory that is scheduled to go on production this year) the world total capacity becomes 124,666 MWh where China will have 42%, the USA 32% and the EU 1.4% of global Li-ion manufacturing capacity. Europe, leading the way in ambition to go 100% electric cars in 22 years time has not matched that ambition with building or planning battery manufacturing plant.
Li-ion Battery Use
Figure 6 This figure combines Tables 3 and 4 from ref 1 to summarise where Li-ion batteries were used in 2016.
Figure 6 (Tables 3 and 4) summarises Li-ion battery use, I presume for the year 2016. Note that China allegedly deployed 94,000 eBuses in 2016 consuming 21.6 GWh of batteries. The total battery capacity consumed 73.7 GWh suggesting that Figure 5 columns 2 and 3 (data for 2015) were on stream come 2016. Columns 1 to 3 totals 76.3 GWh. To summarise Tables 3 and 4 (Figure 6) using the 73.7 GWh total as the base:
- Cellphone 19.0%
- Tablet computer 7.3%
- Laptop computer 10.2%
- Battery electric vehicle 28.8%
- Plugin hybrid EV 5.3%
- eBus China 29.3%
That Chinese eBus figure is eye popping. Electric vehicles (EVs excluding eBus) currently account for 34.1% and this figure looks set to grow substantially. However, we need to recall that batteries consume 35% of total Li production and hence EV batteries currently account for only 12% of total Li supplies. Currently, stationary battery storage does not figure in the statistics, presumably because these numbers are still too low (Figure 8).
Figure 7 Table 11 from ref 1. Dr Kauranen provides a range of future consumption scenarios that are not simple to follow, but this table and Figure 8 provides an overview of a likely scenario.
Figure 4 shows current Li production of 32,500 tonnes per annum with 35% going to batteries, an amount equal to 11,375 tonnes. Figure 7 (above) projects this figure to grow to 103,200 tonnes by 2025 resulting in a 9 fold uplift in battery demand for Li. However, since batteries are not the sole source of demand, this equates to a 3.8 fold uplift in global Li demand in the next 8 years (Figure 7).
Figure 8 A screen capture of Fig 4a from ref 1 shows where future growth in Li demand is expected to arise. The Y-axis is tonnes per annum. While other uses remain constant at 20,700 tonnes per annum, consumer electronics almost doubles from 4,300 to 7,200 tonnes per annum by 2025. But the massive expansion comes from EVs from 7,500 (2016) to 80,000 tonnes per annum by 2025. Stationary storage goes from nothing to 16,000 tonnes per annum, which in the current cultural environment looks like an under-estimate.
Figure 8 illustrates where this demand growth is expected to come from. Pursuit of the Paris accord and recent moves by French and UK governments will certainly underpin this scenario. But will it be possible for global supply to quadruple in such a short period?
James Stafford writing on Zero Hedge:
Right now, lithium isn't even traded as a commodity; rather, it is managed through an oligopoly of three or four major global suppliers who have managed supply and demand for decades. That's why everything is priced on a contract basis.
This has made it difficult for me to find an authoritative source for historic Li prices and there is a tendency for prices to be quoted for different chemical forms (e.g., Li metal and Li carbonate) and in different units. This chart from James Stafford seems as good as any.
Figure 9 Lithium carbonate spot price via James Stafford and Zero Hedge
$14,000 US per metric tonne for Li2CO3 translates to $74,060 US per tonne of metal in early 2016. (Numbers in rounded atomic mass units: Li=7, C=12, O=16, Li2CO3=74, Li2CO3/LiO2 = 74/14=5.29*14,000 = 74,060)
This Seeking Alpha source says this:
The graph above shows April 2017 lithium carbonate contract prices at US$14-15,500/t, and lithium hydroxide at US$16,300/t.
Suggesting that the price rally has cooled for the time being.
The doubling in price since the end of 2015 will send the miners to work, bringing existing resources onto production and exploring for new resources. Past experience tells us that this cycle normally ends in tears for miners and investors but ultimately benefits consumers as low, competitive prices win through. Having said that, the projected demand growth is huge, and supply may not keep up with demand for many years resulting in volatile and sharply higher prices in the years ahead. Escalating commodity price (Li and Co) will blow a chill wind through the embryonic EV industry.
If readers have any red hot investing tips then please share these in comments - battery manufacturers, battery technology companies, Li and Co miners, etc.