Full title: "Stabilization of gamma sulfur at room temperature to enable the use of carbonate electrolyte in Li-S batteries". The paper is from February, but it doesn't seem to have been posted before. TLDR is that with just a new anode made from carbon nanofibers and sulfur, batteries otherwise almost identical to existing Li-Ion batteries would have 2-5 times the energy density, last 3 times as long, and cost 1/3 or less, and be more environmentally benign (not needing cobalt). The breakthrough here is the discovery that carbon nanofibers stabilize gamma-phase sulfur which previously was thought to only exist briefly at high temperatures, and which doesn't form the polysulfides that have been the main problem with Li-S batteries.
Yeah, there have been lots of "breakthrough battery tech" announcements that didn't lead to commercial products, but this one really looks like it may be just the leap forward that will make renewable energy storage a non-problem.
>Yeah, there have been lots of "breakthrough battery tech" announcements that didn't lead to commercial products
Although very important to note in these discussions: a lot of them
did. Yes, in mass production as as part of a whole battery system vs a single lab cell the gains are almost always far smaller, but small gains add up over time. Energy density has dramatically increased over the last few decades [0] as well as cost dropping and those trends together have combined to make the current electrification acceleration possible. There isn't any need to be overly cynical about this stuff, the progress is real and a tipping point was reached a while ago. With the amount of capital for R&D and production pouring in at this point it's not unreasonable to hope for even more.
Energy density, yes; longevity, no. Obviously making batteries that last (much) longer would not be very profitable, especially to the companies who depend on forced obolescence.
This applies less when selling B2B. Auto companies and power companies are GOING to shop based on reliability, and are GOING to notice when your stuff doesn't last as long as it should.
I'm referring to the battery companies, who won't bother to increase longevity beyond the bare minimum necessary to meet a specification.
While I doubt something like the Phoebus Cartel exists for batteries, when they all have little incentive to make them last longer, you won't find much difference between them. Exceptions include military and aerospace where the costs are also many times higher.
I think that in the super fast growing market you don't lose anything by making your product last longer. The demand would be larger than supply for a long time anyway. OTOH, it's good to get an edge over competition by offering a superior product.
And it's not about "big breakthroughs" any more than it is for microprocessors. A bunch of lab changes that in production add .5-5% each add up to 10-20% in a generation add up to 2-3x over a decade. That something works in a lab though is still interesting since it shows the possible ceiling we can work towards. Going from that to something that we can make billions of cheaply entails compromises, but the ultimate envelope still matters.
So this got me curious and I started to look at something a bit more concrete: the history of 18650 cell capacity. Currently the highest capacity 18650 that's widely available is 3500 mAh. The first such cell I found is the LG MJ1 which dates back to 2014: https://cdn.shopify.com/s/files/1/0481/9678/0183/files/lg_mj...
That's a fair bit less than a 3x improvement over 10 years. Now of course this is a rather crude and unscientific measurement. It's also effectively measuring energy density by volume rather than weight, but volume is more often the limiting factor for Li-ion batteries (e.g. smartphones and EVs are volume limited).
Still, I think this shows why people are skeptical about claimed massive improvements in battery tech: it's hard to find clear evidence of it in actual products people can buy.
So far as I can tell Li Ion batteries have gained about a 1.5x charge per mass improvement in the last 10 years. They may have doubled or trebled over a longer time frame though. If there was a big jump due to a particular technology an extra year or two in the time frame, or shift in the time frame might make a big difference.
Just like in finance, most (consumer/laymen) people are looking for a big win when it's been shown over and over that small gains compounded over time always come out ahead.
Mobile phones mostly optimise for Wh/m³ or for batteries more usually stated Wh/L (stored energy by volume), however that tweet is about Wh/kg (stored energy by weight) which is less relevant to mobile phones. “Density” means one thing that has little to do with energy, “energy density” https://en.wikipedia.org/wiki/Energy_density is a slightly better term but it is also misused. “volumetric energy density” versus “gravimetric energy density” are clearer terms (edit:) for usage in conversations like this, although the more standard industry terms afaik are “energy density” for volumetric and “specific energy” for gravimetric.
Apple isn't going to be selecting for mass/energy density. That's a major concern for EVs, but for a phone, you would gladly take double the capacity for 3x the mass.
EV battery capacity hasn't exactly been skyrocketing either. When the Tesla Model S was introduced in 2012 the largest battery option was 85 kWh. For the 2015 model year it got a 90 kWh option and 100 kWh for 2016. Today 100 kWh remains the most you can get.
I can't find any indication that the weight of Tesla batteries has changed significantly since the introduction of the 100 kWh pack. In any case weight savings on Tesla batteries seems to have come primarily from improved packaging rather than better chemistries.
I think while the energy density has doubled the amount observed work from a battery seems constant.
Laptops last a few hours, phones last around a day, etc. We’ve used that density to get thinner and lighter devices with slightly more performance that’s hard to observe from everyday tasks.
The outlier is cars, where they’ve crept up as the density has come up.
Has it really ? I remember the same mAh ratings or slight bumps in phones for many generations. 2016 Samsung galaxy s7 had 3000mah battery - latest ones have like 3700mah ? There have been larger batteries in bigger phones - but I'd like to have a 6000mah battery in a normal form factor.
The only doubling I see is in 10 year period, but phone size grew considerably in that time as well.
And smart watch batteries still suck no matter the price range.
You’re not taking into account the weight of those batteries or their cost as components. A bigger phone might not just have a bigger battery, it might have a bigger cheaper battery with the same capacity. Only looking at the capacity doesn’t tell you anything useful about charge to weight.
> that will make renewable energy storage a non-problem
I recently played around with the data from Germany 2020. I scaled up the existing renewables (only the ones that are scalable - wind and solar, hydro is pretty much maxed out) so that the yearly consumption can be served from 100% renewables. Then I added a battery to carry the overproduction over to the gaps. A 3 TWh battery would still lead to 50 days of empty battery and thus gaps in the power supply.
This is electricity, not Transport and heating. And obviously my approach is simple, no sector coupling effects, no smart grid just scaling up want we have.
3 TWh is 30 million Model S batteries. Even with sodium ion batteries I think this is not really realistic.
I always wondered why the are talking about hydrogen so much, despite the much higher losses you get in that process. Now I get it. I think batteries will be a good solution for single family homes / buildings with solar panel. But the grunt of the grid storage I think will be hydrogen. Especially because during winter, it can produce heat and electricity and the same gas turbines that are already common in German cities.
Either way, it's a big, hard project, but doable if we really want to.
If you overprovision the electricity production you can get away with less battery capacity. It’s a tradeoff and the cost-optimal point is not going to be the one where the production is exactly equal to the demand.
Yes that's why no one is going for 100% currently yearly production match with renewable, it will need to go higher in production :).
Then you'll get times with overproduction with very low prices, this will prop up "power to storage for power later" (liquid, gaz, whatever) with buy low sell high usual market stuff.
Note that France has currently 130 TWh of gaz storage capacity for about 473 TWh of yearly electricity consumption.
Grid storage is different from powering EVs though. You don’t provide a week of grid backup via battery. You do stuff like pump 100 million gallons of water into a dam with a hydro plant ready.
I think water is more efficient then hydrogen even.
Hydrogen is not very efficient, hydro is very efficient. But the energy density for hydro is lower than people think.
m * g * h -> you either need lots of water or lots of height. Best is to have both. You only get that in the Alps and you need some special geoprgaphic features to make it work. Europe is also densely populated and the nature we have left, we want to preserve, so we can't flood all of it.
Do the math, figure out how big of damn we would need to store 3TWh.
My result: at 1000m pump height, you would need 3000 billion liters, which is cube of water with a length 1.5km. Double check this tough, I'm not sure if I did all the conversion correctly
I am not sure there is a universal value for "what people think" about pumped hydro energy density, but your numbers are quite far of: 3TWh at 1000m pump height would require 3e6Wh3600s/h/(1000m1000kg/m3*9.8m/s2) = 1.1e6m3, so a waist-high lake of one square kilometer would do.
For what it is worth, in Europe, Switzerland is doing a lot of pumped hydro powered by surplus French nuclear power in the summer time. Norway has huge amounts of hydro and only pumps a little, but it doesn't matter much: The Scandinavian grid is so well connected, that hydro will ramp down when power is available elsewhere, in effect providing on-demand provider for the whole region.
Sure, but it's very dependant on geography, if you want to do it at scale. We're definitely going to have to adopt a mixed approach, so I'd say it still makes sense to keep the research into hydrogen, batteries, all of it.
> I recently played around with the data from Germany 2020. I scaled up the existing renewables (only the ones that are scalable - wind and solar, hydro is pretty much maxed out) so that the yearly consumption can be served from 100% renewables. Then I added a battery to carry the overproduction over to the gaps. A 3 TWh battery would still lead to 50 days of empty battery and thus gaps in the power supply.
Did you use the approach outlined in https://doi.org/10.1016/j.euroecorev.2018.07.004 ? Because "50 days of empty battery" is vastly different from the results in that work. 3 TWh is roughly 0.5% of German annual consumption and should bring you to at least a 90% renewable scenario, if not to 100%.
90% is pretty close to the 50 days of empty battery. Keep in mind that on empty battery days, you still produce some energy as well.
My approach was very very simple: if production is higher than consumption, store energy. Energy storage is limited, so when the stroage is full, energy is lost. If consumption is higher then production, take the difference from the battery.
I have not considered any smart usage patterns or any advanced concepts at all.
Also keep in mind that this ignores transport an heat. Pure electricity was what I looked at and that is somewhat around one third of the total energy used in Germany.
> 90% is pretty close to the 50 days of empty battery.
That's not quite how I understood "50 days of empty battery", but fair enough. This is not something that can't be solved, though. Even the 90% estimate is the worst case estimate, since there's actually a high likelihood of 3 TWh of storage being sufficient for 100% of renewables in any given German year.
> My approach was very very simple: if production is higher than consumption, store energy. Energy storage is limited, so when the stroage is full, energy is lost. If consumption is higher then production, take the difference from the battery.
That's Sinn's approach ("waste no energy and store everything, no matter the cost"), which yields significantly higher total costs (and storage capacities) than Zerrahn's approach which actually takes costs into consideration and leads to much smaller storage capacities on the basis of throwing energy away or using it opportunistically being actually cheaper than storing it for later grid use.
> Also keep in mind that this ignores transport an heat.
Transport and heat actually simplify things significantly for the grid since charging of EVs in Europe can be shifted around by as much as several days while heat will have been largely solved by the EU's 2010 and 2012 energy efficiency directives.
I'm sure that there is a lot of smart things that can reduce the amount of storage needed. I'm not a researcher in this field, I wanted to do simple but valid calculations that represent a realistic scenario, accepting being more pessimistic to keep it simple.
I also found that increasing the production by 20% does more than increasing the stroage by 10x. I think at 200% renewable production and 2 TWh I ended up with 0 empty battery days in my simple model.
So I believe you when you say that the storage can be smaller. I hope I can also run these numbers myself if I find some time.
I am however also a bit sceptical about all these smart optimizations. It's hard for me to judge hoe much behavioural changes that would need. I'd rather overbuild production and storage than have people reject renewables, because they feel reminded of the DDR when they get their charge-time allotment to be consumed during a certain time only.
Thanks for the names for the approaches, will check that out! I didn't do any research on existing material, I wanted to figure it out on my own first, to get a better feeling for the problem. Often when is start with the papers, I don't actually end up really studying the problem, because I hunt trough (often hard to understand) papers and end up giving up being overwhelmed.
I'm pretty sure that all you need to start with is linear programming and the Aggregate Production Planning model -- which is usually being used for physical manufacturing, but can be adapted to electricity production. Ideally you'd want a MILP solver to make automated decisions to assign certain binary variables (for example whether or not to build certain large-scale storage projects of very specific sizes -- typically PHES, since they're all different), but you can often get by by simply generating multiple models for every combination of binary variables and picking the best one. Similarly you can handle the integration of nuclear power plants, which would be integer variables (for example the number of 1 GWe blocks in operation) -- just generate N models for 0..N-1 nuclear reactors on the grid. I did one such model for the Czech grid some time ago but I'd have to dig it up. The basics are not difficult, though -- just split up the time period into intervals of a suitable size and observe in each time interval that Production_t + Storage_t - Consumption_t - Storage_t+1 = 0. That's what I did, having known production models but having read no papers on the renewable storage subject either. The logic is the same, though. I just got my grid information from https://open-power-system-data.org/ and ran with it.
Exactly, and we think ammonia is even better than hydrogen. It is easier to transport and to store (especially for places without underground salt caverns). See Https://www.airthium.com
The only real reneweable energy storage problem is that we've not built enough generation to really need any storage yet. And so we're shovelling money to authoritarian regimes around the globe which they use to buy politicians and media.
Luckily the EV market alone is enough to drive battery tech for short term storage forward, and green hydrogen for fertilizer is enough to take care of the rest. But, there's no need to wait, we should be a decade ahead on this at least if not for well funded lies holding us back.
Yeah, we do not need any breakthroughs in battery technology for grid stabilization in fixed service. We can do it today with lead-acid batteries if we needed to. Nobody cares about energy density, power density, or mass generally.
To be precise, it's made from hydrogen and the current standard hydrogen is in turn made from methane (and the rest of the methane is dumped into the atmosphere as fossil CO2).
While accurate in a technical chemical feasibility sense, it is misleading to suggest that ammonia is industrially produced from hydrogen. Existing facilities take in methane and output ammonia and CO2. It is true that the chemical reaction used to make ammonia takes hydrogen and not methane, but that fact in isolation ignores a lot of existing infrastructure. Existing ammonia producing plants do the conversion from methane to hydrogen in the same facility that produces the ammonia, making the hydrogen essentially an implementation detail. There are other ways of producing hydrogen industrially, but it is incorrect to think about the hydrogen generating system in isolation when considering existing infrastructure. Hopefully in the future new plants will be built with green hydrogen in mind.
The science says, 2-3x is pretty much a done deal. There are a lot of challenges to productize and that will take time but there are so many independent companies and research groups coming up with different ways to dot this that I feel confident saying at least one of them is probably onto something. I think the real deal is another doubling in energy density to somewhere between 5x and 10x. E.g. solid state batteries might enable that eventually. It's probably a bit further out. 10-20 years at least. But more than probable.
The impact of this for EVs would not necessarily be cars with bigger ranges but much lighter/cheaper cars with similar ranges as current high end models with faster charging times that only need a third of the battery weight. Anything beyond a few hundred miles of range is basically irrational and overkill. Normal people have a bladder range of about 200-300 miles at best (I get uncomfortable way before that) and ought to stop for longer than five minutes when they relief themselves eventually. Perfect opportunity to top up a battery. But most people don't actually drive that far more than a few times a year; if at all.
> But most people don't actually drive that far more than a few times a year; if at all.
Absolutely true, but, until we move to a "usage" rather than "possession" model for cars, this will be the biggest obstacle to massive adoption of EV as the primary car.
At this point people get that they probably only run 100km a day at most, and that an EV would perfectly suit their daily commute (small EVs would be the perfect yellow jacket thing.)
However, everybody has family to visit on the other side of the country twice a year.
The moment you cross the psychological threshold of 1000km on a single charge (or, roughly a full day of highway), the whole game changes.
Would it be better if people commuted by train and ebike, visited their family by trains and had small easy to rent EVs for the days they need a car ? Sure. Will it happen before 70 years of cultural impact of personal car change ? Hard to tell.
> However, everybody has family to visit on the other side of the country twice a year.
Relatively few people drive across the entire country twice per year. Many Americans do drive several hours to visit their family.
It takes basically one trip using a current long range EV (e.g., ~310 miles range) and modern HVDC charging to rid yourself of these concerns. A 9h (~500 mile) trip only requires about an hour of HVDC charging. You'll need to stop for 30 minutes anyway unless you're on some kind of cannonball run.
Except that on the holidays, there will not be enough chargers, so the “30 minute stop” is an ideal that simply cannot be reached under peak holiday conditions. Booking a charger could make things predictable, but it can’t solve the problem of not enough chargers available during peak periods.
Chicken and egg problem: there aren't enough people with EVs yet for it to become a problem. What scares me is the physical limitation of the highway stop area. The _gas_ stations there are already way past capacity on a few strategic days of the year. French people don't like waiting 20m to get gas on August 15th, but at least once they're done, filling up is 5m.
I don't think any TV is going to resist filming infuriated would-be-holliday-goers burning under the sun while they wait an hour or two to fill their batteries. The first summer it arrives will be _very_ hard for the EV industry.
Also, those "highway areas" are owned by... gas companies. So will they put their best will into making sure the EV-holliday experience is top-notch (maybe ramping up charging capacity with portable batteries during the summer ?) Or will they let the experience be hellish, look at the number of EV stagnating, and lobby like hell to move the ICE bans further to the future ?
So people insist. The reality is that people buy the car they can afford rather than the super car they really want. So, whether they will like it or not, most of them will switch over in the next ten years with adequate range. That's why there are so many 40kwh EVs on the road already. They are more than good enough and they get the job done. In the end, spending 50K extra on a car you don't need with some ridiculously oversized battery for a journey you only do a few times per decade is completely irrational. The rational thing would be to split the difference and rent a super car, fly first class, or just man the F*** up and stop to charge for 50 minutes a couple of times the few times per year you have no other choice.
The good news is, it will be technically possible for people to continue to buy cars they absolutely don't need for money they can't afford. Not going to be a problem. It's not really a problem now, really. Manufacturers will be more than happy to take your money.
> The reality is that people buy the car they can afford rather than the super car they really want.
> ... pending 50K extra on a car you don't need with some ridiculously oversized battery for a journey you only do a few times per decade is completely irrational
You seem to assume people are thinking about cars _rationnaly_. I don't know about other countries, but at least in France, this is _not_ the case.
About 30 to 50% of the TV ad space is devoted to cars. The surge of SUVs in the last decade is caused by people buying _entirely irrational_ cars (way too big, way too heavy, with - sorry HN - way too much electronics.)
This is also the perfectly predictable result of marketing incentives (car dealers sell what they're told to sell.) EU has some regulations that _might_ make it unaffordable for car companies to keep the same incentives in the long run - but I don't expect them to play nice.
People will keep buying what their friends, car dealers, auto magazine, music videos, movies, and, mostly, TV ads tell them to buy.
We've been post-homo-economicus in this market for decades.
Now, the "second-hand" market is much more rational - but EVs are not there yet.
It isn't irrational though for people who live in apartments for rentals for example and can not ad charging capabilities. They don't want to charge all the time and if you can get a battery that lasts 1000 miles that would be huge. Also Canadians and others in cold climate are finding range really limited because of the need to run heaters so a battery with much greater range and the ability to run a heater and still have moderately good range in the winter months would absolutely make sense.
It is definitely the weight loss and reduction and materials that would make this a big deal. Imagine cutting the cost of the vehicle by 20-30%. Would be massive.
> Or even EVs capable of 500km, but weighing a tonne less than they do today.
That's not too realistic, a Model 3 battery only weighs half a ton to begin with. So perhaps we could cut the curb weight of the car by 15-20%. I wouldn't complain, but it's not world changing.
Yes a 3x increase would be transformative for general aviation. A lot cheaper to operate. Similar range. Probably will start happening in the next five years. Current models are a bit limited for range. But that should be a solvable problem.
I'm personally thinking in a quite different direction-- that 100 kWh is pretty decent and that the advantage might be that an electric car with advanced batteries could be something quite simple, and something possibly quite cheap.
The transition from a battery weighing 500 kilograms to one weighing 250 is something which I see as allowing electric cars from being highly specialized constructions to something needing much less care in their design. After all, there are ordinary cars with engines weighing 250 kilograms.
Remember that you also need to recharge the car. EV are called environmental friendly, in the sense "they pollute less than ICEs in their life" only if charging is done from renewables.
Multiple studies have shown that, even if an EV is charged from a power grid powered 100% by coal, it still produces about the same CO2 as a 50mpg ICE vehicle. Very few grids are that dirty and contain a mix of cleaner sources like natural gas, nuclear, wind, solar, and hydro. Each cleaner source you add reduce the CO2 output due to that EV. On a typical grid, it only takes about 12K miles to offset the additional energy needed to manufacture that EV.
Unfortunately those studies do not take into account the pollution and energy needed to create actual battery chemistry, just because that happen in the poorest area of the planet and the fact that actual batteries last from 5 years if used much to 8 if used softly, while ICE cars last more than double and materials in them are partially recycled. As I said: remember the TCO pollution, not just the journey. Oh, BTW CO₂ is a problem, but is not the only nor the worst pollutant, it was just chosen because people can't really embrace nor like complexity and is a thing we can reduce, so something useful for economist games, not for science.
Indeed efficiency of a modern ICE engine in a car is 25% max in ideal conditions, while a modern industrial turbine powered by natural gas can get 60%.
In fact even with coal the amount of CO2 will be smaller per distance driven. The efficiency of the latest coal power stations is about 48%. With this and accounting for significant energy losses to refine gasoline from oil the electrical cars are already greener than gasoline or Diesel engines unless one use the energy from very old coal plants with low efficiency.
What efficiencies? Those who demand energivore and polluting oil, gas, coal extraction and processing to a moderately near power plant + the energy you loose in transport + the energy you loose in battery and relevant inverter and intermediate transformers of the energy grid?
An electrical motor is very efficient, a battery system is moderately efficient and combined with all environmental costs to source needed materials, dispose of end-of-life batteries (witch means 5 to 8 years per battery in mean) are huge. Of course pushing refined oil around the world is not efficient either but such infra is already there, while no country in the world can recharge their potential EVs on scale nor we can produce them in sufficient numbers nor we know how to dispose of them.
Costs should be computed in total, not just in the running part. Like cost of trains must take into account the construction of the train network, it's entertainment etc NOT just the energy the train use.
Well-to-wheel, covering every single CO2 molecule involved in producing, transporting, burning and consuming ANY fossil fuel used for electricity production, electric cars are still more efficient and produce less emissions per mile. You could turn over the entire fleet of cars on US roads to electric today, and charging them would not pose a problem for the grid.
Batteries do not reach end of life in 5-8 years. Every EV battery on the US market comes with a minimum of an 8 year 100K mile warranty, and they do not get disposed of the day the warranty ends. They have 10-20 years of usable life in a car, then another 10+ years as stationary storage with reduced capacity. When an EV gets totaled out, the batteries never ever go to a landfill, even 10+ year old ones, as they have so much usable life in them they're still worth thousands of dollars.
For one point of reference, a 10 year old Nissan LEAF EV battery -- which were tiny compared to the batteries that come in new EVs today -- will still have more usable capacity in it than a brand new $11,000 Tesla Powerwall for whole-home battery backup and solar storage.
> For one point of reference, a 10 year old Nissan LEAF EV battery -- which were tiny compared to the batteries that come in new EVs today -- will still have more usable capacity in it than a brand new $11,000 Tesla Powerwall for whole-home battery backup and solar storage.
Before the pandemic sent used cars prices to the moon, there were people buying used Bolts (and probably Leafs, I imagine) to park them permanently just to use the battery capacity for their off-grid home. Apparently it was cost effective, and the Bolt wiring is pretty straightforward to work with.
Also sorry but NO statistic, at least not one public, cover the pollution of lithium mining since that happen in exotic and poor places. Almost no one these days want to publish against the Green New Deal, it's seen like a classic patriotic act by many, unfortunately if you are intellectually honest and really an environmentalist you should know that from advertisement and reality there is a big difference.
100k miles came quickly for anyone who use a car for real, in 4-5 years actually BTW, and while ICEs cars have formally equal guarantees in mileage terms, they can last issueless for around 200k miles, and with not marginal entertainment till 400k. EU commercial vans, bus, trucks are used normally for 2 million km and some last other two in third world countries. So please be intellectually honest. Even carmakers publicly declare that with actual chemistry it's impossible going electric on scale except marginal markets like the (in)famous "glorified golf cart" and few luxury car.
If today all USA cars switch to BEV the USA electricity grid will collapse in few hours and can't recover even with continuous rolling blackouts for years, even here (France) with all NPP operational it will be impossible to recharge the actual ICE magically transformed to EV. That not counting the homes switching to heat pumps.
Just to gives you a realistic dimension this month I've used form grid 151.2kWh and from solar 209.18kWh, in January 460kWh from grid, just 171 from solar, that's in a modern home super-insulated with heat pump heating/cooling/cooking etc. If I count consumed diesel in kWh (around 1000kWh/month), well, I need to triple my small plant. And of course I need two car per head to be used alternatively to recharge them. Now try the same basic calculation for someone who live in an apartment and do not WFH nor that near home. Just to dimension mere energy needs, not counting costs (in natural and economical terms, producing capacity in natural terms etc).
That's the world I live in, but you are describing a world that's totally unlike it. Is "Real Life World" a fictional world you've invented and trademarked with a nonsensical name in order to confuse readers?
> Did you ever have a lithium battery of any kind for more than 10 years?
Yes, I have a 2012 Nissan LEAF which has a 24 kWh lithium battery pack. It is still working perfectly now after 10 years.
I also have a 2018 Nissan LEAF, a 2019 Kia Niro plug-in hybrid, and a 2021 Volkswagen ID.4. Suffice it to say, I have some experience with owning and charging electric cars in the real world.
> 100k miles came quickly for anyone who use a car for real, in 4-5 years actually BTW
The average car in your country reaches 100K miles when it's 14 years old, and in the US when it's over 8 years old.
> Just to gives you a realistic dimension... if I count consumed diesel in kWh (around 1000kWh/month)
You lead with "realistic", then gave a figure equal to driving about 4000 miles in an electric car every month. This is not realistic. Last month my electric cars consumed less than 150 kWh, or the equivalent of plugging in a small single room space heater for 3 hours per day. This fuel cost me about $20.
> If today all USA cars switch to BEV the USA electricity grid will collapse in few hours and can't recover even with continuous rolling blackouts for years
Converting 80% of all passenger vehicles to electric would only add about 10% to total electric consumption. Studies have been done, the grid would be fine.
From what I know, having experimented and seen that's can hardly been. Perhaps you use cars very little so they last longer and traveling just for short trips makes them usable for you even when they can just run 50 miles. But while that's a perfectly legit use case it's far from the mean use cases for all humans...
> Yes, I have a 2012 Nissan LEAF which has a 24 kWh lithium battery pack. It is still working perfectly now after 10 years.
After how many miles/km if I can ask?
> The average car in your country reaches 100K miles when it's 14 years old, and in the US when it's over 8 years old.
The average is meaningless, people living in major cities are a huge percentage of the population and use cars far less since in dense EU cities cars are a needed nightmare because just traveling 20km (~12 miles) demand 30'+ in good days and sometimes, few time per month, can demand hours. While people living outside those major cities travel much, much more. Where I live most travel around 50-60km/day (31-37 miles) around ~10+% travel 100-120km/day (62-75 miles) on average. I have few neighbors who bought Tesla, I see how much their batteries wire out, surely not linearly, but at a rate with their usage, that will make them useless in 8 years, they might last [1] 10+ years but with too little range to be useful. Not only, resale value is essentially zero here. No one want second-hand EVs, no one trust them. Use them to extract the battery to complement a domestic p.v. plant it theoretically possible but impractical since those batteries are not made in easy-to-extract modules, of course their are small cylinders inside, but practically recycle them is costly and complex enough that on scale the price of a new battery will not be much different than of an already dead one, counting it's residual life and fire risk.
> You lead with "realistic", then gave a figure equal to driving about 4000 miles in an electric car every month.
Diesel is rated 10.7 kWh/l (40.5 kWh/us gallons) I use around 60 liters/month (15.85 gal/month). You might counter that electric motors are more efficient BUT in practice they are just a little bit more because:
- I produce energy from p.v. so hopefully recharging out of my micro-plant means from DC (panels) to AC (solar inverter, ~89% efficiency) to DC again (car inverter-charger, probably around 60% efficient) means a big loss of energy;
- battery itself is around 90% efficient
- motors inverter from battery also loose power
Giving a scientific estimate is not easy, you also need to take climate into account, cars HVAC usage, ... but I think estimate a real total efficiency around 15% plus than an equivalent ICE car is realistic enough.
> Last month my electric cars consumed less than 150 kWh
I do not own and EV but I driven some, my mileage declared by their on-board systems was 150-210Wh/km driving normally, witch means ~270-378kWh/month. Assuming a similar consumption for you that means ~33km/day (~20.5 miles/day) witch is very, very little for anyone living outside a city.
> Converting 80% of all passenger vehicles to electric would only add about 10% to total electric consumption. Studies have been done, the grid would be fine.
Seen how much blackouts happen also in developed countries from Texas to Tokyo region just to cite some famous... I doubt that electricity grid can even satisfy actual demands. Since few years, France the most nuclearized country in the world have done very small rolling blackouts to remain at 50Hz. Swiss another big producer of electricity have had issues as Norway due to reduced hydro production. Just in the last years, with not much EV nor extreme weathers. If you imaging a "country wide smart grid" with cars connected to quickly step in if needed... That's a dream. Having a p.v. with lithium storage I see how quick a modern inverter is and the answer is not enough. If you add signaling and relevant vulnerabilities on scale instead of stabilize the big smart grid you crate continuous blackouts. Car's inverter will ripple anytime the load go down, fail to sustain frequency quick enough every time the load ramp up.
Consider a thing: EV actually can recharge super-slowly a 3kW/monophase or slowly at 7kW/monophase or 22kW/triphase (mandatory here for new houses and apartments parking since 2020 or so), actual grid consider around 6kW mean power per household, most apartments are 9kW maximum, most homes 12kW single-phase or if they have electrical heating, SPA, ... maximum 36kW. In some countries like Italy most households have just 3.3kW. And our grids are already in a tough situation now.
Consider that most will charge their car in the evening, no signaling infra exists to tells cars inverter when start to balance the charge in the night, no way to tell cars "when I need the battery full". P.v. start to be popular but so far most parking lot do not have p.v. and even those who have it can't recharge out of it more than few cars. Again no infra is in place and can't be done quickly.
Lithium earth resources are essentially unknown, but current estimates are not much optimistic and we still do not know how to recycle batteries. With the actual chemistry 80% of private vehicles can be EVs around 2050 if we push WFH much, something that start to happen just a little bit, with many resistances and schizophrenic behaviors.
Right, but there are no technological barriers to renewable electricity generation. Not only is it possible with current technology, it's already the cheapest way to generate electricity. The limits on adoption of renewables are all to do with storage.
I have a p.v. system at home, not the maximum power possible but enough for my needs. With an EV I'll have to spend around half the price of the new EV for augmenting my solar part of the mini-plant with a life expectancy of solar inverter around 10 years and for the car from 5 to 8 maximum. After that time car battery got buried somewhere in the third world since we actually have no recycling strategies and I can only recharge from renewable if I do not use the car much and one day yes another no since I produce electricity only during the day.
That not counting the environmental damage provoked by mining for solar panels and batteries. Not counting the fact that we already have issues satisfying electricity demand without much EV on the roads.
Sorry, I'm an environmentalist and an engineer, I know people like dreams, but I also know what we can have, so far the proposed new deal is simply a disaster. I've built my new house well insulated, with actual green standard, because that's a good and doable thing for those who can afford doing so and now I consume far less energy to heat/cool the house, that's good. If gas prices goes up a bit more I might benefit economically from an EV but that's not a green in environmental sense of green, that's dollar green. First we can't rebuild in a decade the 99% of existing buildings to lower energy consumption, it's good going fast in that direction, but we can reach that goal perhaps in a century, and "we" means in the western world only. That's do a good job for us, but it's only a part, even if big enough, of our energy needs. We have industry needs, transportation needs. For industry actual best option is nuclear (constant production per nearly constant demand in developed industrial systems) witch also work for ships, even if is hyper expensive. Some goods can probably be transported by intermittent rail service that move goods only when we have energy, but such commercial-only network is to be built and is a colossal and not flexible at all solution so probably not even worth the investments in environmental terms.
The sole Green New Deal I see theoretically possible is with a mass genocide that kill a large slice of humanity in very short time. With that scarce resources would be less scarce and so we can probably last longer enough to evolve if the reduced number of humans suffice to produce technological advancements witch is a bit uncertain, morality aside. Not counting the little issue of dealing with billions of death in a short time span and resulting social and biological consequences.
Really, try to imaging, to dimension a bit a possible new society and draw your conclusion, I've done that with what I know and that's what I conclude, I'm curious about others opinions. Remember in the game that even our small p.v. plant in modern houses are not born in the backyard and the supply chain and industry behind them need to been able to exists forever if we want to live with such model. Similarly EV does not born in our garage. Just to say I can produce around 25kWh/day in moderately good days, I use around 12kWh for my house in mean, try to design a recharge pattern for an EV. Extend the computation for actual population density and needs. Try to determine how much TWh we consume in gasoline for our cars and how much renewable we need just to recharge a hypothetical equivalent fleet of EVs. Do such basic and approximate math. Add to it, even if it's not needed, the energy we need for industry and for heating/cooling. Really try that instead of dreaming or swallowing advertisements like most do dreaming a miracle car with a solar roof that run autonomously as the owner wish.
I’m curious if they set out to stabilize this particular sulfur allotrope, or if they had more of a serendipitous “hmmm… that’s weird” data analysis moment.
It’s really interesting because there could be so many ways to stabilize interesting new phases that conventional high throughput theoretical screening isn’t going to predict
Apparently it was serendipitous. They were just hoping the carbon fibers would slow down the polysulphite formation, and were surprised when they found stable gamma sulfur[1].
Given that the breakthrough is getting this stable sulfur phase at room temperature that previously wasn't stable below 93C, I'd like to see how well they survive cold temperatures.
"TLDR is that with just a new anode made from carbon nanofibers and sulfur"
The new material they discuss is for the cathode side (not anode). That's side where the lithium ions go when the battery is charged. The cathode of traditional Li-ion batteries is made of metal oxides: Cobalt, Nickel, Manganese (lately trying to get rid of the expensive Cobalt).
The sulfur cathode promises to remove all of the expensive metals in the cathode and make the cathode much more energy dense.
The anode is where the lithium ions go when the battery is being discharged. In traditional batteries the anode is made of carbon fibers (burnt coconut shells). The holy grail of anode is pure lithium anode - there is nothing more energy dense. But the problem with pure lithium is that it tends to get deposited not in an even layer, but in the form of dendrites that eventually poke through the dielectric separator creating a short. In this work they explored lithium anode, but as I understand that was not the main focus. I don't think they solved the dendrites formation problem.
If I’m reading it correctly there’s little significance to the 4000 cycle count; capacity had already stabilized at ~650mAh/g at around 3000 cycles and didn’t drop significantly thereafter.
Am I missing anything obvious that would imply the cycle stability is limited to 4000? From the trends it seems to me that the headline is understated and these might last much longer than 4000 cycles.
I haven't read the whole thing yet, and I'm not really a battery expert, but a battery needs to check really many boxes to be viable in a commercial product.
Among these are: ability to mass produce, resistance to mechanical stresses, safety, peak performance, low auto-discharge, ability to recycle, energy density, charge cycles, efficiency, availability of materials, operational temperature range, storage temperature range.
The latter criteria seem to be well addressed in the paper, the first few not really, or only tangentially (based on a quick ctrl+f for some of the key words).
They successfully used an electrolyte that has been refusing to work for the last 20 years of research on the topic. That's pretty good. Carbonate (ester) electrolytes are unparalleled in stability.
But the abstract says that they're made using sulfur stabilized by[1] carbon nanofibers[1]. Great: now make a million of them.
1: EDIT: nanofibers stabilize the sulfur, but it is not encapsulated. The phrasing "within" in the abstract threw me for a loop.
Here's the step-by-step production they used, it doesn't seem implausible that this could be scaled up in a factory setting. Maybe a modular approach would work, i.e. you might have 100 production lines running simultaneously, each one running the following ~8 steps:
Material synthesis
Synthesis of CNFs. The free-standing CNFs were made by electrospinning. Typically, 10 wt% polyacrylonitrile, was added to DMF and stirred overnight to form a
polymeric solution.
This solution was then loaded into a Becton Dickinson 5 mL syringe with a Luer lock tip and an 18-gauge stainless steel needle (Hamilton Corporation). The syringe with the needle was connected to a NE-400 model syringe pump (New Era Pump Systems, Inc.) to control the feeding rate of the solution.
The grounded aluminum collector was placed 6 in. from the tip of the needle. Electrospinning was performed at room temperature with a relative humidity below 15%. A potential difference of 7–8 KV (Series ES -30 KV, Gamma High Voltage Research, Inc.) was applied between the collector and the tip of the needle. The flow rate of the solution was kept constant at 0.2 mL h−1.
The as-spun nanofibers were collected and stabilized in a convection oven at 280 °C for 6 h in air atmosphere.
The stabilized nanofiber mats were then placed in alumina plates and carbonized in a nitrogen environment up till 900 °C at a ramp rate of 2.5 °C min−1 and then activated under CO2 flow for 1 h in a horizontal tube furnace (MTI. Corp). The furnace was then cooled at 2 °C min−1 until it reached room temperature.
Monoclinic γ-sulfur deposition on CNFs. The free-standing CNF mats were pun-
ched with stainless steel die (φ = 11 mm) and dried at 150 °C overnight under
vacuum.
The CNF discs were then weighed and placed in an in-house developed autoclave (Stainless steel 316) and subjected to 180 °C for 24 h in an oven. The
autoclave consisted of a sulfur reservoir at the bottom and a perforated disk for
placing electrodes at the top. After 24 h the autoclave was cooled to room tem-
perature slowly in a span of 6–8 h.
The electrodes were weighed and transferred in an Argon-filled glove box via overnight room temperature vacuum drying in the antechamber for battery fabrication.
>we stabilize a rare monoclinic γ-sulfur phase within carbon nanofibers
and my mind jumped to "sulfur inside a nanotube again". But the sulfur isn't "within" the carbon nanofibers, as they later specify, and this is new.
>We demonstrate that despite an exposed “un-confined” deposition of
this sulfur phase on the host carbon material, the carbonate-based
battery exhibits high reversible capacity
The sulfur is in contact with the electrolyte. And it works! That's not supposed to happen, but now it does. So the process you described doesn't make carbon nanotubes, it makes nanofibers, which are much simpler (you don't need axial concentric planes).
The nature of science is disappointment, no? Maybe this will work... Nope. Maybe this other way? Nope.
Now and again there's an actual advance and we rejoice, but if you keep an eye on cutting edge stuff it's got to happen a lot that some promising lead leads nowhere.
I'll take "How we end up with scientists running around in circles making the same mistakes over and over." for $500.
If you goal is to learn about things once they're stable and working, I suggest drop reading science journalism from your schedule and just start window shopping at electronics stores.
OK, I was thinking more of summaries of summaries of university news portal announcements on tech sites (which this one is not). I am not against academic publications, scientists should definitely be acquainted with what is happening in their field.
>Electrochemical characterization and post-mortem spectroscopy/
microscopy studies on cycled cells reveal an altered redox
mechanism that reversibly converts monoclinic sulfur to Li2 S
without the formation of intermediate polysulfides for the entire
range of 4000 cycles. The development of unconfined high
loading sulfur cathodes in Li–S batteries employing carbonate-
based electrolytes can revolutionize the field of high energy
density practical batteries.
[italics mine]
These are bold claims. You don't hear that every day. Polysulfides have been an accepted fact of life in Li-S batteries forever.
I think GP's point is that Li-Ion batteries were state of the art 15 years ago, and they still are today. And they aren't much better either, they still degrade from normal use and still expand if you look at them wrong. Except 15 years ago they at least were designed to be easily replaceable.
This is a Li-Ion battery because Lithium isn’t one single technology it’s a family of batteries that use Lithium.
So lithium titanate, lithium cobalt, lithium manganese oxide, nickel cobalt manganese(NCM), nickel cobalt manganese Aluminum(NCMA), lithium iron phosphate(LFP), and now lithium sulfer are all related but they have very different trade offs. 15 years ago Li-Ion battery tech meant worse trade offs so fewer charge cycles, slower charging, less energy per charge, more costly etc etc.
As Xoa posted in another branch, it’s not true that batteries aren’t much better in the last 15 years. Lithium Ion is not a single type of battery but a category of batteries with many types, each with their own benefits and challenges. Numerous changes have been made to the anode and cathodes to improve stability, power capacity, recharge cycles and cost. Recent efforts have focuses on reducing or removing cobalt. Capacity has grown about 5% per year and general stability has improved. I remember reading articles like this years ago where they talked about using LFP cathodes to improve the stability of batteries. Today LFP batteries are used in some Teslas and probably other vehicles.
The cynicism for battery teach research is misplaced. In our lifetimes we have seen real enormous strides in battery technology actually go into production. LFP batteries are now in actual Teslas, provider a cheaper/safer/simpler option for lower range cars that depend on fewer exotic materials/mining. Sodium batteries are in the pipeline. Sulfur batteries may or may not make it too.
the difference between 600 cycles (best 18650) and 4000 cycles is night and day. There's already Lithium Phosphate that is warranted for 6000-8000 cycles. That's 19 years. With 600 cycles your battery will hit the landfield in 2 years. A Lithium Phosphate battery will last 19 years before it has to go to the landfield.
Nobody takes the environmental cost in consideration until you take away the subsidies and look at the real numbers.
I don't think any batteries marketed as phosphate last that long, although they may contain lithium phosphate.
The only commercial batteries I know of that can do that are Lithium Titanate/LTO, that you can mostly only buy on AliExpress and a few specialty places, and are almost unheard of in commercial products(Except maybe some wireless earbuds and capacitor-like applications).
They have like 1/4 the energy density or something, but they are still really amazing. I'd love to see power tools use them, we could run tethered on a USB-PD, charging during idle periods, and not wear out from cycling.
You might be surprised researching the current product offerings. Google "server rack battery". I am off grid with a 5.1kwh server rack lithium phosphate battery. Basically it's all I need. $2200 shipped with accessories, built in BMS with temp sensor and high grade terminals, bluetooth plus RS-485/CANBUS.
Those are pretty impressive! I wonder if they have some special software control to only charge up to 90% or something? I've never seem standalone batteries without smart controls go more than 3k for LFP.
You mean kWh/kg. Gen 3 is supposed to be "out" in 2024, but I doubt that means available to regular people. This article says they expect to start with the space industry, which suggests they'll be pretty expensive: https://www.forbes.com/sites/jamesmorris/2022/04/02/sulfur-b...
Yeah, any new tech will be expensive. In theory, once it hits mass production it should be cheaper than existing li-ion since it doesn't require cobalt or nickel. I'm hoping it works out and hits similar price as today's batteries by 2030. But I'm probably too optimistic.
> I have to wonder what economic chance new battery types have given the massive investment in current tech.
Current tech tends to improve at a steady pace.
Taking a new tech to production takes some years (somewhere in the 5 to 20 years range); processes need to refined and scaled, suppliers sourced, people trained, recycling must be addressed etc.
So new tech only stands a real chance if you can convince investors that its advantage is big enough that it still dominates when it comes to market.
Let's say current batteries improve 7% per year (number made up, but not totally unrealistic). If the new technology needs 10 years to get to market, the current technology improves by a factor of 1.07*10 = 1.97 in mean time. So a 2x advantage along any axis (density, charge cycles, price, availability of materials etc.) is not a solid bet for an investor. A 5x advantage likely is.
If course you'd want advantages on multiple axes to really make a solid bet.
There's lots of segments for battery chemistries to be used, so they just need to beat out smaller segments to establish commercial viability.
EV batteries also aren't a monoculture. They are even mixing battery types in EV packs to achieve different range/cost aspects.
If there's room for Sodium Ion, LFP, and the various cobalt/nickel chemistries, if there's air, sea, and long-haul modes plus grid storage, home battery storage, plus all the mobile and tools, there's plenty of segments.
Yeah, there have been lots of "breakthrough battery tech" announcements that didn't lead to commercial products, but this one really looks like it may be just the leap forward that will make renewable energy storage a non-problem.