As an aside, we're not actually short of lithium. What we're short of is current sources, so this is about exploiting a mechanism to find new sources cheaply, to sell to developers who want to make a killing on lithium futures. Because right now, there's actually quite a lot of lithium out there, ready to be extracted.
A Tesla battery uses 63kg. There is estimated to be around 50MT (megatonnes) of Lithium in the supply chain and known deposits. If we fermi-maths this around, call it 50kg per car, thats 20 per Tonne, so we have enough in known deposits for 1 billion tesla cars.
Sure, thats less than one car and one powerwall per person on the planet. So in some sense, we know we'll need more.
But, whilst I applaud this use of satellites, please don't mistake this for news because earth-sensing sats have been used for mining research since before tesla was a thing. As the article says: its only new using it to search for lithium.
Mining guy here. I have done plenty of feasibility study works for lithium projects over past few years and in general;
There is a lot of supply coming online from hard rock lithium sources in the next 12-24 months, especially from Australian spodumene sources.
Demand side is heavily from the Chinese electric car market. Smaller cars, only 5-6kg max LCE per car. All other demand is a small percentage of what Chinese demand is estimated to be.
And yes you are right, we have been using satellite imagery for decades, it’s just a lot easier now.
My understanding is it is cobalt that is the problematic element in short supply. We don't have many supplies outside of the DRC and that has a lot of problems.
Biggest source by far is the DRC, but there are plenty of other sources. The problem is they are difficult to process and beneficiation is costly. But with rising prices, this should become less of an issue.
Dorothy Dunnetts 'nicolo' series of bodice-ripper historical fiction has its roots in the discovery of Tolfa (alum) mines and the impact on the dye industry. The clue to location being (amongst other things) holly and a differential preference for alum rich soil. I guess in some sense remote sensing for vegetation and then focussed analysis has deep deep roots.
Interesting that lithium sources are going to rock. I thought the saltpan stuff in the Atacama was a price beater for years to come. I also believe the various rare earth dependencies are determined not to get caught behind China as a sole source.
Brine sources are massive, cheaper sources, but take years to come online. The problem was there were about 3-4 major players in Lithium (it was actually a side product for them) and they didn't see the demand rising as quickly as it did.
This allowed for a massive price increase, which made it feasible for hard-rock miners to come in. With a decent project, you could go from finding the resource to production in 12-18 months, as the ore is normally close to the surface, and processing for export is relatively simple.
So the next 12 months a lot of these projects are starting to beging production, while the brine players play catch up.
Atacama will make it's operators very rich, however it is not practical for them to raise extraction rate to where they could serve the entire market. So, in addition to them, a lot of rock-based supply needs to be developed.
For reference, at one point people thought there was going to be a "uranium boom" during the cold war era. So a bunch of prospectors went around the US looking to stake out claims to uranium deposits. They thought they were going to beat everybody to the punch of a late 1800s style gold rush. Only problem: They soon found out that uranium occurs so frequently in the crust that there's no economic incentive to monopolize deposits.
In the link, you'll see that lithium is about 10x as common as uranium. I'm also sure I'm not fully aware of all the factors here - but just based on that information I think we're fine for lithium reserves.
Those maths don't actually make me feel too great.
Considering the number of devices (cars, tablets, phones, laptops, heated clothes, robots, etc) in development or production that depend on lithium for batteries, not to mention any other pharmaceutical and scientific uses, it seems like only 1 billion Teslas worth isn't gonna cut it if we're planning to keep up with the pace of technology acceleration, penetration and turnover.
Oh, and I just remembered that Tesla is planning to make EV long-haul trucks... what's the math work out to for how many of those can be made? I think I saw a video that said the battery would need to be tremendous in order to carry feasible loads.
What about if we did the math on how much lithium the average western consumer is expected to use over the next 10 years given current device turnover and expected increase of lithium-based device ownership?
Of course this wouldn't even factor in working government and business use cases and the consumption based on new startups making destined-to-fail products.
Lithium in batteries is highly recyclable, which helps quite a lot. It’s not a pure consumable like oil. And none of this takes into account the possibility of novel chemistry taking over at some point.
Since lithium is one of the lightest elements, it doesn't seem possible to create batteries that will have the same power:weight ratio. MinutePhysics explains it better than I ever could, basically all the options require lithium:
Regarding recycling, you're right, lithium-ion batteries are 100% recyclable. However, the economics don't make it feasible, so it's not going to be done until the that equation changes:
> Recycled lithium is as much as five times the cost of lithium produced from the least costly brine based process. It is not competitive for recycling companies to extract lithium from slag, or competitive for the OEMs to buy at higher price points from recycling companies. Though lithium is 100% recyclable, currently, recycled lithium reports to the slag and is currently used for non-automotive purposes, such as construction, or sold in the open-markets. However, with the increasing number of EVs entering the market in the future and with a significant supply crunch, recycling is expected to be an important factor for consideration in effective material supply for battery production.
So once the price of lithium increases 500%, recycling will become feasible.
All I'm saying is that my intuition on the numbers isn't as optimistic as others. There are diminishing returns on all our natural resources once we pass the peak price for extraction. Lithium is just one of those resources that I feel we're being relatively careless about.
In a world where economics favored long-term sustainability over short-term profits we would be increasing the price ahead of time to slow extraction and increase incentives for recycling the invention of lithium-sulfur or lithium-air batteries now, while there is still easily accessible lithium in the ground.
7 billion people don’t need 7 billion cars. Many dense cities don’t require cars, and many households don’t need one car per person. Our multitude of battery devices use far less lithium than a single car. Yes the math sucks but it’s not as bad as it looks.
I grew up in Europe. This is an exaggeration. Heck, even when I lived in London the location of my house vs my office made it pretty much impossible to commute except by car.
I don’t know if London is the best example. It wasn’t really designed at all over the last 2000 years, it just grew and grew and infrastructure had to grow around it.
Personally I found Italian cities like Rome and Ravenna to be cozy and walkable. Polish cities too were also quite accessible without cars.
The plan is to use satellites already in orbit to detect and map geological and botanical features that might betray the presence of subterranean lithium. Though satellite prospecting of this sort has been employed before, to look for metals such as gold and copper, using it to search for lithium is new.
The searchers are not searching blind. They know, from mining records dating from the mid-1800s, that there is lithium in Cornwall’s rocks. Those records tell of underground springs containing salts of lithium—at that time quite a recently discovered element.
TLDR; Li ore prices have doubled in past 10 years. How do you search for it from space? satellite cameras, both optical and infra-red, and also satellite-borne radar, to look for mineral formations caused by hot liquids reacting with existing rock, and for rock fractures that could act as channels for lithium-bearing brine. They will, as well, record anomalies in vegetation that might be the result of lithium-rich soils, or of hot springs that might contain the element.
Because we still have no good micro payment system. So payments are slow (enter creditcard numbers, paypal credentials) and expensive (whatever you use, transactions have a USD 0.30 miniumum plus a percentage...).
I would gladly pay .10 cents for each article that I read until I reach 3 dollars a month which is what they currently offer but I think that they like the repeatable revenue. I am guessing that they are betting that people will buy the economist and then just have it keep re-subscribing.
Because otherwise you get incredibly terrible advertising. Although from my experience with paywalls, you still get that same bad advertising experience for 5 bucks a month.
A Tesla battery uses 63kg. There is estimated to be around 50MT (megatonnes) of Lithium in the supply chain and known deposits. If we fermi-maths this around, call it 50kg per car, thats 20 per Tonne, so we have enough in known deposits for 1 billion tesla cars.
Sure, thats less than one car and one powerwall per person on the planet. So in some sense, we know we'll need more.
But, whilst I applaud this use of satellites, please don't mistake this for news because earth-sensing sats have been used for mining research since before tesla was a thing. As the article says: its only new using it to search for lithium.