If size were not an issue could reliable storage batteries be built cost effectively.
The rule of thumb for project management:
Cheap, good, fast. Pick two.
Reliable storage = Good
Cost effective = Cheap
Therefore, time is the variable you have decided to allocate the most flexibility to achieve your goal.
So the answer to your question is Yes. Someday, we will have such a battery.
Do you mean for grid storage?
They’re already being built cost-effectively for some applications. Namely, as a replacement for gas turbine peaker plants and rotating mass. They’re effective for storage on the milliseconds to hours scale.
For longer durations, we’ll need cheaper materials. It’s helpful that grid storage isn’t as weight/size constrained, but the limit isn’t infinite. The batteries still have to be trucked in and so on. The battery casing isn’t free.
I’m optimistic about sodium-ion batteries. Very cheap materials and only marginally less capacity than lithium-ion. But they need more development.
I’m really quite surprised at the scale at which Li-ion batteries are being deployed. First electric cars, then big trucks and whole-house power backup systems, then grid storage as a replacement for gas-fired peaker plants.
I would have thought that at the scale of grid storage, other technologies would be cheaper and more practical. At least some of our nuclear power plants around here use an array of stationary jet engines as an emergency power source for the grid, for example, and those can be powered directly by hydrogen. Rolls Royce already has a working prototype actually intended as a potential aircraft engine. I think hydrogen has a lot of potential as a source of much higher energy density than any conventional chemical batteries.
And I think the problem with hydrogen is precisely the very low density. Even when liquefied it only has a density of 0.07 g/cm3, and you lose one third of the energy contained in hydrogen to liquefy it and you don’t get that back when you use it. Storing liquid hydrogen is hard, either you keep letting some evaporate to keep it cool, which is bad for long term storage but OK for immediate use or you have to keep sinking energy into the storage system to keep it cool.
Ammonia is probably easier to produce, store and hande, even if it is problematic too.
Aside from the practical advantages (weight, etc.), there’s also a huge knowledge base behind li-ion battery systems, coming from the auto world, driving the tech forward and the costs down. It’s no coincidence that Tesla as their fingers in both pies.
You can also just go onto their site and order a “Megapack”. Their 4-hour pack is $8.34M for 19,600 kWh, or $425/kWh. The spot price of li-ion cells is getting to around $100/kWh, so obviously you’re paying for a lot of equipment and such.
But they’re using the LFP chemistry and probably in a fairly mild configuration, and might get 10,000 cycles out of it. So a cycle only costs $0.04/kWh.
While power sometimes gets that cheap, at other times that might be a great price. It’s really cheap compared to typical gas peaker plants. So you charge with cheap solar before the duck curve (peak A/C demand) or at night with cheap wind power, and discharge it in peak hours where the price might be $0.50/kWh or higher. That $0.04/kWh premium isn’t much in that case.
All that said, people are looking into all kinds of alternative technologies, but li-ion is going to be hard to beat due to the volumes, and because it’s starting to become a relatively small portion of the total system cost. Halving the cost of the cells wouldn’t have much effect on the Megapack prices even at the same density, and if the density got worse, the non-cell pack costs might go up enough that there’s no overall savings.
We’re starting to see the same effect with solar. There are various technologies for super-cheap solar, but at lower efficiency. The problem is that it’s so cheap to start with that the packaging, installation, etc. costs dominate. You’re better off with a relatively expensive (but high efficiency) cell that packs more into the same area.
And lots of that is thanks to the mobile phone. The need for high power density batteries that could be recharged many times was always around, but all of the sudden there was demand at a scale that supported doing the research and development to get those batteries.
And “battery” has more meaning than a lithium cell–pumped hydro is a battery and one we should be using a lot more.
I think maybe Edison nickel-iron batteries may be an answer. They’ve been largely displaced by batteries that are more powerful per mass or volume, but perhaps for the OP purposes that isn’t a problem. They last a long time and take many recharges. Though, I am afraid they may bleed off their charge faster than many would like, so not sure.
Yep. It’s neat how massive scale in one industry can unlock the capital investment and R&D necessary for a technology to be useful in another industry. Mobile phones are responsible for a few of these–another example is the MEMS accelerometers, gyroscopes, magnetometers, and barometers that are so ubiquitous now.
This is what I was actually thinking about. Cheaper metals, easier to dispose of.
Indeed, I think they’d all get recycled.
Nickel is expensive and better used in combination with lithium (and manganese and aluminum) for a higher energy density. Also, NiFe batteries will outgas hydrogen and have other negative qualities. Their robustness isn’t nearly as useful in an age of electronic charge controllers that can manage the charge/discharge curves very precisely.
Lithium-iron-phosphate is almost perfect, except for the lithium. Using sodium instead would be fantastic if it didn’t damage the energy density too much.
You need significant vertical elevation, a decent plot of land up on top of the high terrain, and lots of unallocated water. Those things are not in large supply in much of the USA. Where there are suitable sites I agree completely they can make a lot of sense. If there is excess generating capacity nearby and the spare grid capacity to move it.
It’s not obvious to me that there are enough of those sites, and especially not in the flat and arid parts of the USA where most of the population growth is occurring.
As with any hydro project, whatever is immediately downhill needs to be something that’s willing to have that inherently dangerous pile of potential energy up there on the hilltop. See
for an example of what happens when they have an oopsie.
IMO, we could do a lot better in using non-pumped hydroelectric. Dam generators are not always sized much beyond average use. But if they’re instead sized for peak use, they become much more like a battery pack. It might seem wasteful to have generators that only run for a few hours a day for a few months of the year, but the lake is acting as a giant battery in that case, and probably still cheaper than doing the same with electrochemical batteries or otherwise.
Well, I’m in Montana, which has ample wind power and topographical relief. Gordon Butte. I drive by there often–lots of wind turbines, almost no people, lots (sort of) of water. This is true of may places in the intermountain west, including the deserts, and it doesn’t seem like you would need too many water rights–your main losses will be evaporation.