Hacker News Reader

Comment by yrro | Hacker News Reader

According to https://www.epectec.com/batteries/cell-comparison.html, 160 Wh/kg is about the same density as Li-po and Li-ion. This battery chemistry is attractive in that it's made from common materials & is more stable/safer than Lithium. The press release doesn't say, so I assume it's not competitive in energy density per litre so I assume not.

Wikipedia has a comparison table at https://en.wikipedia.org/wiki/Sodium-ion_battery#Comparison but no idea how accurate/up to date it is.

View on HN

It's made of sodium and iron, which together make up about 8% of the earth's crust, so yes, they chemistry is made of really common materials. By contrast lithium makes up about 0.002%.

Misleading way of looking at it. Lithium and sodium are not the major cost (or weight, or volume) inputs to making batteries, and crustal occurrence is very distantly related to cost. We mine things from places with 100x-1mx higher concentrations than natural. Water concentrates lithium into brines and clays for us. Sodium's low density causes it to create massive domes underground that are extremely recoverable. In contrast many metals aren't naturally concentrated.

Lithium batteries aren't made of lithium. They're made of nickel- or iron, or manganese, or cobalt. In iron and manganese batteries the #1 price factor is the manufacturing- the energy, solvents, and machinery used to deposit materials onto film.

Likewise sodium batteries are not made of sodium. There's 13x more iron in them than sodium. There may also be large amounts of manganese or vanadium. The cost of manufacturing is also higher per kWh.

Lithium batteries are made of lithium to a point that there is a lithium supply bottleneck if we want to use lithium-ion batteries as the base chemistry for the green transition.

It will take time to mass produce things regardless, but I imagine Sodium has far fewer bottlenecks.

The lithium bottleneck idea comes from people who go "looking at the current lithium supply we would run short if we instantly started producing 10x as many batteries as we do today", completely ignoring how markets work.

It's great if we can get a chemistry that avoids the need for lithium, but it won't be a showstopper if we don't.

There is no lithium bottleneck, we merely haven't even bothered to catalog all the lithium that is easily accessible.

Yes and: No long term shortage with occasional short term crunches. Mostly due to lag time bringing new supplies online.

Like with every commodity market.

Lithium is still part of the cost of producing batteries and their production still requires a large number of lithium mining operations. Having a ton of lithium mines around isn't exactly what's destroying the planet, but having fewer would be more better.

This sort of misses the point of sodium ion batteries though, no? One of the main objections to lithium ion batteries is the need for cobalt because of how it's sourced through "artisanal mining" in Africa.

LFP has no cobalt.

And can be regularly charged to 100%, and lasts for a greater number of charge cycles.

Northvolt’s original factory in Skellefteå is near (well, ‘Arctic near’: 460 km along an existing rail line) Kiruna, one of the largest Iron mines in the world, so that’s one of the two material supply safe.

Industrial Sodium is made with electrolysis of sea salt; the factory is next to the Gulf of Bothany and has abundant (wind and hydro) power, so the other material supply is safe.

It wasn’t hard anywhere, but it’s straightforward in that particular case.

This got me thinking - the salinity of the water in the bothnian bay is very low (seems to be about 1/10th of ocean water). Wouldn't that effect electrolysis?

Possibly—but if that’s a concern, you can also get some from the North Atlantic.

https://www.reddit.com/r/coolguides/s/Co4zeAhcmT

Iron is on top. Lithium is one up from the bottom left.

strange how calcium and sodium are omitted

Sodium is not "mined" per se, as, you know, just get some sea water

Calcium is, but maybe because it's not processed as most metals it is not included in the graph

At least half of all NaCl that's used world wide is mined from salt mines. For many places in the world it's not feasible to rely on solar evaporation of sea water. Using other energy sources to evaporate sea salt is not cost effective and many places have large salt deposits.

https://salt-partners.com/pdf/Santorini2006Paper.pdf

I am aware of this, but mining (or salt water processing) for the specific extraction of sodium metal from NaCl or others is really small

Mined salt is probably more valuable as table salt (and cattle feed) than as source of metallic Na

Would it be feasible to use the sodium from desalination wastewater?

desalination wastewater contains a number of other chemicals used in the desalination process (e.g. pH adjusters, coagulants and flocculants, antiscalants, dispersants, biocides, and reducing chemicals)

So the next question: would those contaminants significantly degrade the performance of the battery?

I mean... pH adjusters would definitely significantly alter things. The other major problem (I'd guess) is just the health implications of working with toxic wastewater. Is it safe?

For context, as of 2019, we produced enough of this "brine" to cover Florida with 30 centimeters of brine every year. That means, as a whole, desalination plants actually produce even more toxic wastewater than they do clean drinking water.

As a result figuring out ways we could utilize this _product_ ("byproduct" feels like the wrong term here considering it's the primary thing produced) is a major area of interest

I looked up your claim that desalination plants produce "toxic wastewater" and I found nothing to support it. The output appears to be simple concentrated ocean water, and that's it.

Can you cite your claim?

I'm more interested in what you possibly looked up that made it hard to find...

> Furthermore, chemicals such as biocides, surface−active agents, anti-scale additives, and solid residues from filter backflushing may be present in the effluent discharge on a continuous or periodic basis, posing a risk to the environment

> Brine effluents from RO desalination plants not only have a high salt content but typically also contain compounds from the desalination process, such as phosphonate-based antiscalants and ferric (or alum) sulphate-based coagulants

[1] https://www.frontiersin.org/articles/10.3389/fmars.2022.8451...

[2] https://www.sciencedirect.com/science/article/abs/pii/S00431...

> However, concern also exists regarding the use and release of toxic anti-foulants and anti-scalants to maintain plant infrastructure

[3] https://www.sciencedirect.com/science/article/abs/pii/S00431...

> The super-salty substance is made even more toxic by the chemicals used in the desalination process, researchers reported in the journal Science of the Total Environment.

> Copper and chlorine, for example, are both commonly used.

[4] https://phys.org/news/2019-01-brine-highlights-toxic-problem...

Didn't gasoline start out as a byproduct of kerosene production?

Sure but it's alternative uses were already known. It just so happened that a world-altering invention (the consumer automobile) came along to dramatically raise already existing demand for it. There is currently no demand/use for desalination brine. For every "this byproduct is actually useful" story there's likely 10 byproducts that simply stay byproducts. Still, it's urgent we figure out something to do with it since it's damaging our ocean ecosystems

They can be removed in purification steps.

Go from the super high salinity brine through to crude salt, then chloralkali process to get sodium (which can be cleaned up) and chlorine gas (industrially useful).

Sodium is also mined for sure. There's a reason there are many expressions about salt mines :-)

I was editing after doing some basic research so sorry for making it look like you're repeating my comment :)

Lithium is element #3 on the periodic table so it's very simple and should be universally abundant. Literally in the universe, unfortunately not on earth.

That’s not how it works! Lithium is a fuel that gets used up by stars immediately whenever it might be produced in trace amounts. Unlike hydrogen, lithium wasn’t produced in the Big Bang. So most of the lithium that remains in the universe is produced outside of the cores of stars through the interaction of cosmic rays with other matter. Needless to say, that’s not a very common interaction (relatively speaking).

Now if you look at how larger stars operate (the CNO cycle [1]) you’ll see that it matches up with the higher relative abundance of carbon, nitrogen, and oxygen in the universe. Lithium, beryllium, and boron get “skipped over” in a sense.

Furthermore, if you look at a graph of the relative abundance of all elements, you’ll note that odd-numbered elements are less abundant than even (with the exceptions of hydrogen and beryllium). This is called the Oddo-Harkins rule [2] and it may also be playing a role.

Edit: I should also add that the third major process in stars, triple-α [3], involves the fusion of three helium-4 nuclei into one carbon-12 nucleus. This occurs in older stars that have exhausted most of their hydrogen fuel and so have built up a large core of “inert” helium. When their outward pressure from hydrogen fusion is no longer high enough to withstand gravity, they reach the much higher pressures and temperatures needed for triple-α fusion. Unfortunately for the lithium industry, there’s no chance of producing lithium this way since it is skipped over on the way to carbon.

[1] https://en.wikipedia.org/wiki/CNO_cycle

[2] https://en.wikipedia.org/wiki/Oddo%E2%80%93Harkins_rule

[3] https://en.wikipedia.org/wiki/Triple-alpha_process

Nitpick: Lithium was produced in the Big Bang, though in a ratio of something like one per billion compared to H production.

https://en.wikipedia.org/wiki/Cosmological_lithium_problem

There is a lot less Lithium in the universe than you might expect being element #3

There is also a lot less Lithium in the universe than our models predict:

https://en.wikipedia.org/wiki/Cosmological_lithium_problem

This stinks of bad science. All of the observations come from stars. "Older stars seem to have less lithium than they should, and some younger stars have much more."

"BBC Science Focus wrote in 2023 that "recent research seems to completely discount" such theories; the magazine held that mainstream lithium nucleosynthesis calculations are probably correct."

I am unconvinced.

Helium is no 2 , and that too is pretty scarce on earth , but again helium is a very light gas and simply shoots out of the atmosphere eventually, Why is lithium rare

Lithium is not very universally abundant.

You cannot estimate abundance by atomic number like that. The big bang produced mostly hydrogen and helium, with traces of lithium and beryllium. The elements heavier than that are mostly produced by stars, and the physics of fusion have a massive impact on what elements, specifically, get made. Free protons join together to become helium-4 much more readily than any other fusion process, meaning that by the time heavier things start forming, the raw material is entirely ⁴He.

This means that things that are easily made of ⁴He are dramatically more common than anything else, making the most common isotopes after ⁴He oxygen-16 (4 alphas), carbon-12 (3 alphas, less common than oxygen because it's less stable and easily picks up another alpha), neon-20 (5 alphas), and iron-56 (14 alphas to nickel-56 which immediately decays twice through β+ to produce ⁵⁶Fe). Iron is so high up above all the other intermediate steps, because it's the last stop: In heavy enough stars, the entire core converts to iron, and reactions past that are energy-consuming, not energy-producing, so after that the star collapses.

Lithium is not on any of the major stellar nucleosynthesis pathways, which means it's only produced by exceptional processes, making it roughly as universally abundant as the other stuff that is made by exceptional processes, like scandium or gallium or zirconium. But none of that matters, because:

Lithium is abundant and easy to extract in the earth's crust.

While there's not that much of it up there, there's plenty easy to extract down here, because it's so light and likes forming light compounds, meaning that a huge proportion of all the lithium of all the rocks that came together to form the earth is reachable to us. Lithium is not rare. Any statement about lithium batteries that bemoans the scarcity of lithium is doubly confused: Firstly, because lithium is simply not scarce. Secondly, because lithium is such a tiny portion of the battery, that despite being in the name, only a small fraction of the materials cost is lithium.

Lithium price has had a few big spikes because mining is a very high-capital industry where spinning up projects is measured in years, if not decades, and we suddenly started using a lot more lithium in ~2010. Accordingly, the price has spiked from the ~$5k per ton (which is roughly in the same ballpark typical cost of extraction, where any abundant mineral prices end up at), to the heights of $37k per ton last year. Even at this high price, lithium was not even the most expensive material component in most lithium batteries, because typically only 1-3% of the battery's weight is lithium.

But these prices won't last, because having the price of a commodity so high above the cost of extraction means that new mining projects are spinning up.

Re: Lithium as a resource:

USGS (2021):

    Five mineral operations in Australia, two brine operations each in Argentina and Chile, and two brine and one mineral operation in China accounted for the majority of world lithium production.

    Owing to overproduction and decreased prices, several established lithium operations postponed capacity expansion plans. Junior mining operations in Australia and Canada ceased production altogether.

USGS (2023):

    Six mineral operations in Australia, one mineral tailings operation in Brazil, two brine operations each in Argentina and Chile, and three mineral and two brine operations in China accounted for the majority of world lithium production.

    Additionally, smaller operations in Brazil, Canada, China, Portugal, the United States, and Zimbabwe also contributed to world lithium production.

    Owing to the rapid increase in demand and prices of lithium in 2022, established lithium operations worldwide increased or were in the process of increasing production capacity.

Sources:

* https://pubs.usgs.gov/periodicals/mcs2021/mcs2021-lithium.pd...

* https://pubs.usgs.gov/periodicals/mcs2023/mcs2023.pdf

Bonus British Geo. Soc. Global Li Map: https://www2.bgs.ac.uk/mineralsuk/download/global_critical_m...

Also from the 2023 USGS periodical:

    Lithium supply security has become a top priority for technology companies in Asia, Europe, and North America. Strategic alliances and joint ventures among technology companies and exploration companies continued to be established to ensure a reliable, diversified supply of lithium for battery suppliers and vehicle manufacturers. Brine-based lithium sources were in various stages of development or exploration in Argentina, Bolivia, Chile, China, and the United States; mineral-based lithium sources were in various stages of development or exploration in Australia, Austria, Brazil, Canada, China, Congo (Kinshasa), Czechia, Ethiopia, Finland, Germany, Ghana, Kazakhstan, Mali, Namibia, Nigeria, Peru, Portugal, Russia, Serbia, Spain, Thailand, the United States, and Zimbabwe; lithium-clay sources were in various stages of development or exploration in Mexico and the United States.

Sure, there's two or three pages there IIRC.

If you want to go in depth, though, you can always hit:

https://www.spglobal.com/marketintelligence/en/campaigns/met...

Thank you for a fascinating comment. I learned new stuff from it; I appreciate you and your expertise!

Thank you for this explanation.

> because lithium is such a tiny portion of the battery

Is this why recycling it is so difficult?

Lithium is extremely abundant on earth. Unless we start launching it into space, or start building up after covering the surface with buildings and roads, we’re not going to run out.

Lithium production capacity is scarce however, since it’s a mostly useless element unless you’re building batteries out of it.

Anyway, once cities realize that they need to stop taking water from rivers, we should be able to skim quite a bit of lithium from desalination plant waste water.

I mean in that case Hydrogen fuel cells are clearly the future, just as soon as we manage to make our gravity well irrelevant.

> the chemistry is made of really common materials

I don't want to sound like a conspiray theorist, but something tells me the really big actors (like states) only want materials that they can control the suply of.

> sound like a conspiray theorist

Well you do?

It won't work as long as there's a roughly equal alternative that's cheaper/easier to produce. Free market will win here.

There's no way one state can force another state (aside from war) to manufacture something a particular way. It's like if I controlled the world's timber supply and said Canada must produce houses out of timber and not, say, concrete. Canada's gonna go produce using concrete unless I somehow make my timber price competitive.

> Free market will win here

Think batteries and nuclear fusion.

Extremely hard stuff, not easy to pick apples.

State actors can absolutely influence the fields for decades by choosing to fund certain approaches that lend themseves to centralisation.

> It's like if I controlled the world's timber supply and said Canada must produce houses out of timber

This is quite naive - in fact we do this all the time

* IMF provides loans to developing countries on the condition that they dont have 'socialist' policies

* EU bailouts for Greece/aspain/etc. was given on the condition of sale of state assets and doing other things

* The worlds ship insurance industry is run in London. Nuclear powered contsiner ships are faster, cheaper, and better in every way. Good luck insuring them. Running them without insurance is. illegal

*'non-tariff barriers' - i.e. free trade negotiations - are all about aligning countries on how they manufacture/insure/regulate things like cars. Guess which econony gets the bigger say.

Russia was forced to adopt Eu standards for petrol quality and engine emissions standards in 2,000's and they still follow

IMF provides loans and expects them to be repaid, having rules on government spending attached to the loans is... reasonable. Not to mention, that IMF loans are typically bailouts of governments that overspent.

Same goes with the EU bailouts, but PIGS countries were already in a compact with the rest of the Eurozone. Not to mention, that governments should not own things that can go bust and drag a budget under water.

As shown lately with Russian oil sales - it's absolutely possible to insure ships somewhere else, other than Lloyd's of London.

You're right I was too flippant with my language. I was only thinking about the US and its strategic desires.

> IMF provides loans to developing countries on the condition that they dont have 'socialist' policies

Because these 'socialist' policies are usually the reason why these countries need IMF loans.

China already has GWH scale sodium battery plants. So if THEY have been trying to suppress it THEY aren't doing very well.

This announcement is about an improvement in energy density made possible by $Bs being invested to allow sodium batteries to become more competitive with lithium.

There are also other battery chemistries being rolled out. Iron based ones seem particularly promising for stationary storage.

That's actually the official policy in many cases. For instance, the EU is funding research in li-ion recycling so that it could create a "circular economy" with imports only there to make up for material lost during processing, as e.g. the car market is largely saturated, so the expectation is that demand won't grow.

Yeah, but that's not a conspiracy or even something negative. It's just common sense for a country/group of countries.

OP made it sound like the Evil Corporate Overlords are conspiring to hold us back from achieving battery freedumb.

> It's just common sense for a country/group of countries.

Conspiracies are just common sense for a group of people; but that's exactly my point. A group doesn't have to meet at night by the torch light in black robes and decide upon secretive actions to further their own interests.

I don’t think control goes that far. China definitely thinks that way, but I doubt the western governments do.

If they can; abundant is the next best thing.

Any country without an expeditionary military force (about 187 of them) likes the resources they have. Ab abundant is great unless you have a known military adversary with extra-territorial ambition (that’s three countries).

Transistors are the most valuable thing we can produce per kilogram and sell easily. We had many processes over the years, but we settled on making them from sillicon. I.e. sand.

Think about it :)

As always, whoever has the biggest guns will control this resource as well.

If someome resists, they will end up just like anyone who opposed the US's quest to take other nations oil.

As Donald used to say:

"Take the oil, then get out". They took the oil and stayed.

Which countries did the US take oil from? Any data around number of barrels ect?

They're probably referencing Iraq and I'm not sure it was raw resource extraction as it was so much removal of a competitor (whatever the name of the Iraqi state oil company was) and more than that, enabling sales of equipment, consulting, etc. It wasn't as simple as some 1800s colonialism, it was advanced wealth extraction worthy of the 21st century.

Ok. So the US spent $3 trillion[0] on a war in Iraq to get some consulting contracts from a country with a GDP of $36 billion[1]? And didn't invade Saudi Arabia, which actually has oil? How much wealth do you estimate the US extacted from the war?

[0]https://www.hks.harvard.edu/publications/true-cost-iraq-war-...

[1]https://data.worldbank.org/indicator/NY.GDP.MKTP.CD?end=2001...

While I don't lend credence to it being that simple, it's worth noting that the people making that profit aren't the ones paying for it, and the ones paying for it aren't using their own money.

The Europeans were starting to loosen Iraqi oil sanctions and develop the fields before the second war.

The US often does stuff that costs taxpayers trillions so that the people bribing congress can make billions.

PFAS, Canadian lumber sanctions and oxycontin are three recent examples.

>The Europeans were starting to loosen Iraqi oil sanctions

Any links I can read about this? I'm open to the idea that suppressing Iraq's oil industry was the main objective of the war. I don't like claims about "the US's quest to take other nations oil" being that it never happened either in Iraq or even Iran. At least when I ask for a source I can never get one. To me the wars in Iraq and Afghanistan were mainly about projecting power, not oil. Certainly not Afghanistan because there is little to no oil there in the first place. Even regarding Iraq it is OPEC that sets the price and I doubt they would let Iraq greatly reduce the market price. It would have to be as you say: people with connections using the US's power to suppress competition. Many people online, however, seem to have the idea that US foriegn policy dictates collecting oil and that the US is stealing trillions of dollars of oil from various third world countries. I think the US gains a lot more from war to project power. Iraq for the most part today is a US ally. And if we are looking for people who would gain from the war it would more likely be Lockheed than Exxon. Lastly, there is no reason to say that US oil companies staged the war exclusively. It is possible that eg. SA were also involved or the main initiators.

I'm sorry, but Saudis were the biggest losers in removing Saddam.

Removal of Saddam removed one of the biggest adversaries of Iran. Now Iranian Revolutionary Guard can freely move from Tehran to Beirut and support the rebels in Yemen.

The US spent a ton of money from the US public so that a few US individuals can stuff their pockets. Corruption.

You literally contradicted your own original statement.

Those individuals used the US gov to do it, see Cheney. Read up about Leopold and Congo.

Wasn't Iraq about to switch away from trading the oil in USD ?

But let me flip your argument -- why did the US invade Iraq?

The US has/had troops stationed in Saudi Arabia and Saudi Arabia was in the US's pocket at that time (it might have reversed since then).

>Wasn't Iraq about to switch away from trading the oil in USD ?

I don't know. Do you have a source for that claim? Preferably from before the war started. (Later is fine too.)

>But let me flip your argument -- why did the US invade Iraq?

This doesn't flip the argument. That would only be the case if not being able to explain the war meant that it therefore was started for oil. These are not two sides of a coin. Simmilarly I wouldn't say that if we can't explain Iraq then it must have been to find aliens.

>The US has/had troops stationed in Saudi Arabia and Saudi Arabia was in the US's pocket at that time

And what percent of the oil profits did the US get? If the claim is that the US will invade for oil then being in the US's pocket, whatever that means, is irrelevant. But I think you answered your own question here. The US likes having countries "in it's pocket." It wants to station troops in other countries. These are legitimate -- here I don't reffer to moral legitimacy -- national objectives. Skimming contracts or suppressing oil fields is not a legitimate national objective. If there are groups in the US (eg. Exxon) that are so powerfull they can make the US go to war without a single national objective achieved then they could simply take the $3 trillion directly. I think your point about the petrodollar is interesting, although I'd like to see some more evidence.

Iraq, Syria comes to mind. Yes, the US is still in Syria. They would have tried that shit with Venezuela, but Ruskies got there first.

If it's cheaper than lithium but not significntly smaller, at least it'll be more scalable and affordable for e.g. energy grid or home battery applications.

It also has a vastly superior safety profile, also meaning is easier and safer to construct. It does not have the overheating problems of lithium batteries.

I've long dreamt of being able to to have a battery buried under the cellar floor. Size and weight wouldn't matter. Lifetime and safety would be quite important.

Is the lifespan of sodium-ion better than Li-ion?

They claim 2'000 cycles for their current 18650s, which, I believe, is about twice that of li-ions?

"Lithium Ion" encompasses a lot of chemistries. LFP, which is what is most competitive with sodium ion, has a cycle range of 3000-10000.

Depends on the anode

With the moves in places like California to curtail the value prospect of net metering for solar (particularly during peak hours), home storage is becoming more and more important. But I don't especially like the idea of big lithium batteries around the house ... particularly because I live in a flood zone.

While I am very pleased to see these developments away from Lithium I do think your estimates for Li-po and Li-ion are off by a few generations of batteries.

Lipo can be 200+ /kg density and Li Ion can be 250+ in current, commercially produced, generations of battery cells.

I'm not a pro so anyone feel free to correct me.

Not sure about the exact numbers, but your sentiment is basically accurate.

This is an article about the Northvolt news by a German journalist specialized on battery technology (in German): https://www.golem.de/news/akkutechnik-northvolt-und-altris-e...

He says that 160 Wh/kg is in the ballpark of LFP batteries from five years ago. It is, however, about the same as the sodium batteries announced by CATL in 2021.

This would be a way better article about the topic, than the press release by Northvolt, if it wouldn't be in german.

Frank Wunderlich-Pfeiffer should consider writing in english, I love his expertise and clarity of writing.

Volume also matters.

Yes and: Their anode uses "hard carbon", not graphite. Apparently without sacrificing energy density.

This is huge. HUGE.

China dominates the graphite market and now has export controls.

IIRC, most current Li and Sodium batteries use graphite anodes of some kind. Northvote's use of hard carbon may prove to be an amazing cost and derisking advantage.

I know nothing about their novel Prussian White cathode.

I eagerly await the expert analysis of Northvote's anode and cathode.

Fun fact, that is why PR uses 160 Wh/Kg as its current "best in class". The Northvolt release is a bit cagey on what they actually have today for sale but it seems like they are going into production so that is a good thing. "Whole house" energy store (think PowerWall types of products) are getting good traction and grid scale batteries have changed some folks minds about what is "good enough" (you don't care if it is maximally dense if you can spread it out over an acre or three). So I would expect them to push for this sort of application first.

The current market need for large local battery store for EV chargers is apparently one of the limiting factors in deploying new chargers, delivering spot excess demand can be provided by either onsite diesel generators (like some rest areas in California are doing now) or a battery bank. The latter is preferable for energy efficiency and maintenance reasons.

> so I assume it's not competitive in energy density per litre so I assume not.

Their competitive argument is a fast charging time with a low impact on the life the battery pack, with a full charge under 10 minutes and about 2'000 cycles. They also have a good available power and capacity at 20C discharge rates.

Sodium is extremely plentiful, while lithium is not.

I had thought that this was not a huge win, as lithium is fairly cheap, and not a large portion of the overall cost of a battery. However, my research taught me I was incorrect.

Lithium is worth about $40k per tonne, or $40 per kg. A Tesla power wall 2 is about 150kg, if half of that is lithium, then the lithium alone is worth $2.3k. Powerball costs about $9.5k, so the lithium is a fair portion of the cost.

https://www.thisoldhouse.com/solar-alternative-energy/review...

https://www.statista.com/statistics/606350/battery-grade-lit...

Note, I know raw lithium carbonate is not stuck directly into a battery, just spitballing with the little bit of learning I just did.

Lithium is cheap because the externalities of the environmental damage it causes is not accounted for in the pricing. It's a highly exploitative resource which has destructive impacts on local bacterial ecosystems, human communities, and water availability.

Some articles, if you are interested:

https://www.sciencedirect.com/science/article/abs/pii/S09626... https://www.euractiv.com/section/energy-environment/news/fac...

It's not even comparable to sodium, which is abundant practically everywhere.

> Lithium is cheap because the externalities of the environmental damage it causes is not accounted for in the pricing.

Like every other raw resource we use.

Let's not flatten it. Different materials have different externalities. And are available in different places with different levels of human rights and environmental protections

Sodium is better than lithium in that respect. But both are MUCH better than hydrocarbons.

The amount you need for driving a car for 3 years is several kg vs tonnes. And you can recycle the battery but you can't recycle the oil you burned.

That's why I'm not particularly harsh on lithium externalities. Let's get the low-hanging fruits first before we focus on nuances.

Lithium and sodium are both easily mined from sea water.

Seawater contains less than 1ppm of lithium (compared to 300-7k ppm in brine). There are zero commercial facilities to produce lithium from sea salt. It's not even a notable byproduct from other seawater-based processing facilities

Now compare to fossil fuels.

> A Tesla power wall 2 is about 150kg, if half of that is lithium,

This estimate is very far off.

1% is closer.

That's a electric car battery, optimized for mobility: fast charge & low self-discharge, maximum density allowed by the projected lifetime, custom form factor, heat and cold resistant, vibration resistant and mechanically sturdy etc.

When you think of an application like grid connected energy storage, most of those performance metrics are irrelevant, and the only thing that really matters is cell cost per total energy stored and delivered during its lifetime. We will likely see something over-engineered and simplified to maximize cycle count and minimize cost, leading to a much larger raw material consumption, at the expense of density - the cell is not going anywhere.

So the ability to use dirty cheap ingredients is a game changer for the grid storage market.

>A Tesla power wall 2 is about 150kg, if half of that is lithium

an order of magnitude less. 30KWh is just about 3kg of lithium in theory. On practice it would be about 7-10% of the weight of the battery.

Do you have a source for that? I don't doubt what you write, but I would love to learn more.

realistically the cobalt based Li-Ion can reach ~250Wh/kg (and they are better than the LiFePO4). So 3kg of cobalt based li-ion would be below 1kW/h

Also right now about 70% of all lithium comes from only two places (Chile, Australia). Iron and sodium are pretty much everywhere so this potentially eliminates at least one supply bottleneck

> eliminates at least one supply bottleneck

The CIA wants to know your location /s . I know this kind of joke is not appreciated on HN (for good reason), but one has to ponder of the implication of cheap/dense energy/storage and what big actors like governments, big corporations would think about not being able to effectively control energy production/storage/distribution.

If you are the Chilean or Australien government you would maybe be unhappy about moving away from Lithium. Most other governments would love it e.g. Europe doesn't have much Lithium (or at least not a lot that is easy enough to extract to make it profitable). The EU and european governments already try to rely less on foreign supply chains, especially since they relied so heavily on Russia for gas and now have to scramble to find other sources.

> Europe doesn't have much Lithium

The USA doesn't care if there's Xium inside the USA.

Xium just has to be in a few places and it has to be moved across the globe, transacted in USD and guarded by the US Navy.

Big corporations will not invest if they can't create a moat.

Maybe the corporations of those countries. But Australian and Chilean citizens both loathe the environmental and health impacts of these industries. Especially those that live in or around the "sacrifice zones" of these industries

At least in Europe, governments are going to great efforts and expense to decentralise and decarbonise the production, storage and distribution of energy. The implication is that as well as producing energy through renewables close to where it is used, it can be cheaply and sustainably stored there as well.

> lithium is fairly cheap

For now. But more importantly, there are sovereignty problems to considered in case things get worse in the future. And the quality and usability of the lithium substrate varies quite a bit between suppliers, with the better ones, for now, coming from the less "attractive" suppliers.

lithium is not the issue at all for Li-Ion.

Price of lithium jumped 6x in 2020-2022: https://www.iea.org/reports/global-ev-outlook-2023/trends-in...

but is now down by 75+% since last November peak (and still declining). I suspect that we won't see another price spike like last year's for quite some time.

Keep in mind sodium ion and LFP are much safer and don't require nearly as much cooling and management systems as nickel-cobalt chemistries

So at the PACK level of energy density, which is really all that matters, sodium ion and LFP close much of the gap with nickel-cobalt.

So spitballing here, an NMC chemistry at 240 wk/kg at the CELL level will lose about 20+% ore of density per weight for cooling and safety, so that they will be effectively 160 wh/kg at the PACK level.

Most CATL literature has LFP and sodium ion at 90-95% at the pack level with "cell-to-pack" which bypasses modules and other intermediate packaging.

So if 240 wh/kg NMC chemistry is actually 160 wh/kg at PACK level, and this sodium ion is 160 wh/kg but about 150 wh/kg at PACK level, well then you see the real power of these chemistries.

If the pack level 160-180 wh/kg equates to a 400 mile car, then 140-160 wh/kg sodium ion at pack level equates to a 300+ mile car.

300 miles means a really good city car. It means you can probably do a 50-100 mile PHEV car pretty cheap. It means cheap, limit-is-number-of-factories scaling of EV battery supply.

Sodium ion is supposed to be 40$ or less bill of materials per kw-hr compared to 80-100 for NMC and about 50-70 for LFP. And it should probably drop from there in the long run.

It also means that EVs beat ICEs on drivetrain cost, possibly by a significant margin, which might translate to a 4000$ + price difference from an ICE. Combined with theoretically cheaper maintenance and "fuel" costs, this should translate to an EV cost advantage that people simply won't be able to overlook.

Personally I think there should be an overall "carbon externality charge" of $5000 on a new ICE as well, or something that scales with the carbon inefficiency of the vehicle (so a bigass suburban assault vehicle is like $10000).

Also, note that the roadmap for batteries of CATL, a lot like the roadmap for future nodes in semiconductors so take it with a grain of salt as to when they realize the goals, is for 200 wh/kg sodium ion and 240-260 wh/kg LFP. With superior cell-to-pack density, that should mean a 400 mile car for sodium ion, and a 500 mile car for LFP.

Now, hopefully in 5-10 years we get lithium-sulfur and sodium-sulfur that are AT LEAST 50% more dense with similar materials costs. Then you get to shrink the battery to make the EV even cheaper.

So the revolution is coming, in my opinion. And this isn't just a gee-whiz a faster pc for my Overwatch. This is "future survival of humanity in the balance". We NEED to decarbonize transportation, and we NEED cheap batteries for alternative energy grid storage. The development of these technologies is preservation-of-humanity level of importance, and high density sodium ion chemistries are a major major step towards that because of all the economic and practical levels/needs/requirements they meet/exceed.

> "carbon externality charge" of $5000 on a new ICE as well, or something that scales with the carbon inefficiency of the vehicle

Your whole writeup was inspiring and gives me more hope for the future. This part, though, I'm angry about. I'm angry that we don't already have this legislation in some form. I'm sure it will be fought tooth & nail by the big auto manufacturers, but we should do it anyway. Maybe we could tack on higher penalties for anyone caught 'rolling coal', too.

New Zealand has a scheme (soon to expire with the change in government) for this.

Low efficiency vehicles are taxed on import, and the money raised is returned as rebates on high efficiency vehicles.

A Ford Ranger might attract the full fee, a new t Nissan leaf would get the full credit. A small ICE car attracts a smaller fee. Hybrids are given a smaller credit.

The exact amount of credit varied over time as the fees gathered changed.

The sort of obvious way is to slap on a decent carbon tax on fuels. But of course that is fought tooth and nail by a lot on entrenched interests.

Even here in ostensibly progressive Europe, populist parties are riding on "Cheap gas!!!".

It should be a bit more than $5000. However, prepare to be even angrier:

Burning a gallon of gas generates 20lbs of CO2 (most of the weight is the O2), so 100 gallons produces a ton. Direct air carbon capture should cost roughly $100 per ton at scale, so the fee should be $1/gallon of gasoline (either at vehicle purchase or at the pump).

That’s completely affordable and lower than current gasoline taxes in many places.

If we made that one change (and funneled the revenue into carbon capture) existing ICE cars could be carbon negative in 5-10 years, and, as we phased them out (because EVs are just better) we’d have a clear path to pre-industrial atmospheric CO2.

With what process can you capture and permanently store carbon from the atmosphere for that price?

The EU mandates a minimum €0.36/L excise tax on gasoline, which is about $1.49/gallon. In practice many large countries like Germany, France, Italy already levy more than $3/gallon. But they certainly don't funnel the revenue into carbon capture.

Agree with all. Especially this:

> So the revolution is coming, in my opinion.

Yes and: The nascent thermal batteries (box of hot rocks) and advanced geothermal power generation are just now crossing the chasm.

Both tech stacks have been proven, have financing, and initial customers.

And now they're jumping on to the cost learning curve.

Roughly, thermal tech today is where solar and batteries were in the 2000s.

The will be huge because 1/2 of energy consumption ends up as heat. So skip all the middle steps.

I'm optimistic that the rise in geothermal will be astronomical once it gets off the ground. Both advanced and more simple conventional for district heating.

We're at most 10 years from the confidence in oil and gas investments being completely shattered. A lot of the investors and engineers will seek out opportunities where they can apply their competence. Geothermal is a good fit. Whoever captures the market first will have the most to gain, so once they see it's even remotely possible there will be a race.

I suspect politicians in countries with oil/gas-development in northern regions will start subsidizing this as well, both to attract voters from workers in that sector, and to help them establish a new competitive industry that they can replace their oil and gas exports with.

And how strange to rate a storage battery 'per kilogram'. It's just sitting there, on the grid, storing. The weight is entirely irrelevant.

The interesting number for stationary storage is, Wh per $. I wonder where how they compare on that (relevant) measure?

CATL's Sodium Ion is claimed to be 1/3 the price of Li-ion. It is a lot cheaper per KWH but also a little bigger than LiPho which itself is quite a bit bigger than Li-ion.

I haven't seen it that cheap yet, its got new tech prices at the moment for cells on aliexpress.

> LiPho

LiPho? Are you thinking of LiFePO4, aka LFP?

Yes. Sorry Long Covid brain fog I keep thinking there is a H in that I know there isn't because I have 3 of them in my house!

Reply on news.ycombinator.com