Smaller, cheaper flow batteries throw out decades-old designs
spectrum.ieee.org
spectrum.ieee.org
I was surprised by this! Since this is not commercialised on a broad scale, I would have thought that the cost is nowhere near competitive. Twice seems very close. How close are we to introducing flow-batteries at a competitive price?
So if this research pans out it sounds like very close. This isn’t a new concept and it’s mature so the probability of continuing savings is less than traditional battery technologies (eg there’s new kinds of batteries coming out that are cheaper and more energy dense than lithium ion). However, at grid scale this may have additional advantages beyond pure cost that make it attractive still.
But the big cost reductions don't happen because an idea is mature, they occur because they're being produced at large enough scale that lots of people have spent time and energy on all of the thousand tiny things that individually reduce the cost. The stuff in this article is borderline just industrial design and they're talking a 3x change in footprint?
There has to be lots of other low hanging fruit... we're still improving lithium battery electrodes and each improvement really does improve them. The same will be true for flow batteries if they ever get made at enough scale.
Not unreasonable. The details are always in the research paper after all.
Increasing the size of the tanks to increase the energy storage capacity, without increasing output power would be... useless.
Extreme example: a huge expensive 100 MWh battery that you can only draw a max power of 100 W from = useless.
Basically the statement is lying through omission.
Isn't the relevant measurement energy density (Wh/L) rather than power (W/L)? I guess there's some limit on the power generated by a given volume of the battery, but in practice it seems like the main question is how big the tank has to be per MWH/KWH, rather than how big the power module has to be to convert that back to energy over the duration of the discharge period.
> Liu and colleagues focused on redesigning the power module
A flow battery separates the power portion (the electronics, electrodes, pumps, etc) from the energy portion (tanks, fluid). Which is not so amazing for a car or a UPS where you want tens of minutes to a few hours of charge or discharge time, but is potentially great for grid use, where a discharge time of days to months is useful.
I have no idea whether there are containment or environmental challenges that make this especially hard.
It would seemingly also allow you to carry around 50 miles of juice when you are running around town and bump up to 400 miles of juice when you are taking a road trip. This really only makes sense if the car is able to recharge the juice.
Maybe a new hybrid model (not gas electric hybrid) could emerge. The base (say 50 mile) capacity is solidish state like we have today and a flow battery range extender could be filled with fluid when the extended range is needed. This would make it so the car wouldn’t need to charge the fluid, if that helps.
motor <-> lithium <-> flow battery
the low current from the flow battery could be continuously augmenting and replenishing the lithium ion battery
or even:
supercapacitor <-> lithium <-> flow battery
And it's easy to imagine retrofitting a gas station to all of a sudden have charged electrolyte instead of gas, but I'm sure that has all sorts of additional complexities that I know nothing about... ;)
In a UPS the tanks and valves and pumps are new points of failure, as the other responder states. In a car they are many times more complicated than electric vehicles, yes, but still less complicated than ICE vehicles by far. Your heater has more moving parts. Hell, the emissions control gear is probably more complex than the entire drivetrain of the electric vehicle.
A Li-Ion battery can charge and discharge in something like an hour. If you double the number of cells in a battery or installation, you can store twice as much energy and you can draw twice as much power. (You don’t have to provision twice as much power circuitry, but you do need to purchase that extra power worth of electrodes in the cells.)
If you’re building something for which a roughly predetermined power to energy ratio (i.e. hours of use at a reasonable rate of discharge for the technology), this is fine. Similarly, if overprovisioning power or energy is not a problem for cost or weight, also fine.
But for grid use, wide differences in the duration of energy storage for different purposes can make sense. And a flow battery can separately provision power and energy. This gives a possible cost benefit to flow batteries.
It also have lower energy density so you pumping 600l of new electrolyte on the station might still give you only 200km of range
or maybe a locomotive, which is less sensitive to weight and cargo capacity?
For locomotives ? I'd imagine the simplest solution would be to have "battery wagon" ; get on the station, replace the battery wagon, and be on your merry way while it is recharging in the station waiting for next train to pick it up, at least for the long haul stuff. But capacity density still matters, if flow batteries are being close and cheaper then it makes sense, but if you need to pay slighty more per KWh but get 2x the capactity... thats 2x the range and less maintenance.
Or, you know, just electric rail. It already works fine in many places. I wonder if making some hybrid solution with electric rails also being covered by solar panels (basically so same maintenance crew can manage both) would make sense
So it all depends to what level of density it gets. If it is same capacity per kg but lower power there are still many places it can be used, if it is much lower it stops being sensible for moving stuff. And the price would need to be significantly below the normal batteries for anyone to even bother.
Trains maybe, but surely the money is better spent on overhead wire? Even if tunnels/bridges/etc. can't take it you're better with just the power-buffer battery and no flow battery.
And rail is far cheaper to operate once built per unit capacity than trucking. If you don't inflate it with endless feasibility studies and contracts that get paid regardless of output the initial building costs are about double a 4 lane highway, but it carries a lot more and maintenance is lower.
Not sure I know the use case for a duration of that length. It's possible today with pumped hydro, but I don't think it's used in that mode.
The DoE's "long duration storage earthshot" effort is loking for 10+ hours, which makes more sense: https://www.energy.gov/eere/long-duration-storage-shot
Chemical storage could be a big win when a mountain isn't readily available. Or the land is too expensive to use for power storage.
There are other options, of course - new nuclear plants, importing renewable power from countries with better weather, huge numbers of wind turbines, and so on - but cost-effective long-term power storage would address some issues if it was available.
Better long distance transmission infrastructure is a much better solution to this.
Winds are stronger in winter time.
Depending on your risk tolerance you overbuild to reduce that. Same as with all types of energy production.
Why store winter heat as work when you could just store it as heat or chemical energy?
All you need is a hole, some plastic sheeting, an element, and some pipes to store many GWh. Alternatively one of many cheap phase change materials, thermochemical stores, or some low grade sand unsuitable for construction.
Lump those together, halve it for the magic of heat pumps but then add some back because winter will be above average use, call it 30kWh/day. Three months of winter is ~90 days, and ~30 million households, is storage of 30,000 x 90 x 30million = 81,000,000,000,000; eighty one trillion Watt-hours.
A classic (non-EV) car battery stores ~1kWh, so eighty one billion of them to store that kind of power. Thousands per household. Even to store ten percent of it is hundreds of car batteries per household which is still unfeasible, and not going to get you through winter.
Working your example - google tells me that zinc-iodine electrolyte gets you about 200 Wh per liter. Therefore you need about 400 million cubic meters of storage for the capacity you suggest. To estimate the cost of this, I looked at reservoirs. The largest drinking reservoirs in the world are in Qatar, where they have built 5 tanks of 436,000 cubic meters each. Therefore we'd need about 200 of those facilities. The cost was around 5 billion dollars, so our total cost would be around a trillion dollars.
This is obviously a lot! But not unimaginable - the government borrowed over 300 billion pounds in the 2020 fiscal year alone just to pay for Covid. In practice you'd need considerably less than 90 days of continuous power, because the wind does blow and the sun does shine even in winter. All you really need is tanks to buffer the difference in renewable capacity between winter and summer, which is certainly not 100%. And the system can be built up incrementally over a long period of time, and still yield value - we don't need to spend a trillion dollars all at once. And in the worst case - you can erode the fraction of load taken on by renewables with nuclear (doesn't look so expensive now eh?).
Obviously take all these numbers with a grain of salt, this is a back-of-the-envelope calculation built on another back-of-the-envelope calculation (in particular, I haven't included the cost of the electrolyte). The point is merely to show that it's not orders of magnitude outside our ability.
World GDP is only around a hundred trillion, and we're talking of maybe ten trillion of it to build a heating system for Western Europe, for very little financial return for any investors.
Practically, we've been struggling to build one nuclear power plant (Hinkley C) for twenty years. I think it sounds like it is orders of magnitude outside our ability.
I understand it as analogous to a ICE, where you have the fuel and the motor. But with flow batteries you can also go in reverse to create fuel from energy. The research featured here is about the motor but if you need a different energy density you have to change the type of fuel since that's how the energy is just stored in huge tanks.
If you want the storage density, you can look by the battery chemistry. But I'm not sure the article got this one right, or else the researchers didn't actually work on flow batteries and that part is only speculation.
>our SBMT cell shows peak charge and discharge power densities of 1,322 W/Lcell and 306.1 W/Lcell, respectively, compared with average charge and discharge power densities of <60 W/Lcell and 45 W/Lcell, respectively, of conventional planar flow battery cells.
So, it's not per volume of electrolyte, but per volume of the entire cell (housing included).
I'll believe it when someone actually produced one at cost efficient prices.
And 200 wh/kg sodium ion is on the "roadmap" for the next few years. They've been generally about a year behind their roadmaps, but those projections aren't some pray-for-invention thing.
And sodium-sulfur? Hoo boy, if they get those going it's like 500 wh/kg, although last I saw they were using graphene to get the prototypes work, so we'll see.
Are other comments correct? $800/kwhr for flow batteries? Yikes.
Lithium is not the problem. There's plentiful lithium. And I'll believe the prices when they actually start selling it. I've heard "expected" prices many times before and they always turn out false.
Also still to be resolved is the lifetime of that kwh price. Again unproven with many wild claims.
Flow batteries are the thorium nuclear reactors of the battery world.
> And sodium-sulfur? Hoo boy, if they get those going it's like 500 wh/kg, although last I saw they were using graphene to get the prototypes work, so we'll see.
There's many magical claims in the world of batteries. Reserve your judgement until someone actually produces one. It's not just energy density but also power density and cell lifetime that needs to be considered.
Sodium ion is going into production at CATL and Gotion. It's GOING to be in EVs. Sodium Ion is happening. 160 wh/kg, 3000-6000 cycles. Stable. Good temperature range. I don't know any other way to emphasize this.
https://www.catl.com/en/news/665.html
This isn't a research paper, a clickbait headline. They are ramping up production for 2023. Not prototype cells looking for investors, etc. It's not a far flung announcement. This is "hey who wants to buy a shitton of these for a 300 mile range car?" call for sales.
And Lithium may be not really scarce, but you still need to find the sources, develop new sources, extract it, etc. Sodium is far more abundant than lithium. Like, 1000x more. As much as you want.
If it was January 2022 your comment would have weight. It doesn't now.
To emphasize, this means the 300 mile range (real not WLTP) sodium ion car that is fundamentally cheaper than an ICE. City cars for billions of people. No cobalt constraints, no nickel constraints, no waiting on lithium development. You're just waiting for the companies to make more production lines as fast as they can.
This is the beginning of the end for the ICE. There's no maybes or what-ifs or whatabout is there enough of this or child slave miners of that or we have to wait on South American government Z to allow Y.
Simply that this marks the point in history where you don't need to worry about feasibility of the switchover from ICE for 90% of commuter traffic. It's just a matter of time and scale, and ICE staring down the barrel of a drivetrain it can't economically compete with and still likely will drop another 50% in cost in the next ten years.
There are a lot of low-cost vehicles there that are designed around battery swapping, and one luxury car maker, Nio, also offers it. Nio is setting up shop in Europe and already has swap stations in Norway where EV penetration is high.
Your inability to imagine anything other than the laziest most spoiled american's habits (even when alternatives are easier) doesn't invalidate the huge space of other solutions.
If all of those don't work for some reason, then just treating it like an ICE and usingthe 250kW fast chager is only an extra 15 minutes once every week or two.
This is how things work for poors. They don't have access to lots of credit, so a car that costs $15K less is good, and they are used to whatever a tank of gas costs, and will pay that. Again, this is the mass market, the next billion drivers/families. You're well down the income ladder at that point.
More importantly, this is the kind of thing which government funding could cover. Just the healthcare savings along from getting rid of ICEs would pay for a lot of it.
India seems like an ideal market in that a cheap EV with a smallish sodium-ion battery represents a huge step up in mobility for many crores of Indian families. That's less true in Europe and North America.
1. https://cen.acs.org/materials/energy-storage/Reliance-buys-s...
So some behemoth Model S or X sized car in the US that needs a 100kwh pack to get 300+ miles needs 20kwh hours to get a kei car 150 miles.
And of course e-bikes, e-mopeds, e-scooters, etc.
Although it is still early days, with less than a GW and a TWh total installed at customer sites.
Sodium-sulfur's advantage over lithium is long lifetime. (4500 cycles at 100% depth-of-discharge and at least 15 years calendar life claimed).
0. https://www.ngk-insulators.com/en/product/nas.html
1. https://www.energy-storage.news/ngks-nas-sodium-sulfur-grid-...
2. https://www.basf.com/global/en/media/news-releases/2019/11/p...
No, this is completely uncompetitive, so batteries as costly as that would only be used for space applications, for which flow batteries are probably unsuited.
$800 per kW (not kWh) is plausible as a capital cost.
One reason to use flow batteries in grid applications is extremely long lifetime compared to Li-ion (tens of thousands of charge-discharge cycles, multiple decades of calendar life). Grid engineers are used to thinking in terms of the next forty years.