Check out the post I made on the Battery Board about the new Form Energy battery using iron&air process which is in R&D.
Check out the post I made on the Battery Board about the new Form Energy battery using iron&air process which is in R&D.
It’s just a matter of degree and perspective. Energy will be provided by a mix of continuous nuclear and intermittent renewables and pooled into a single pipeline. Despatchability will occur by a mix of batteries for short term gaps or in times of excess by diverting some of the energy to other uses, one of which I believe will be the production of hydrogen. For emergencies natural gas plants are held in reserve. Under such a system the nuclear reactors would be on all the time at their most efficient output.
Long-term grid-scale energy storage has long been searched for as kind of the holy grail for renewables.
I think the answer will come from hydrogen, with nuclear playing major role, hence my investment in Bloom energy. Thanks for this opportunity to pump the stock.
From what I’ve read I don’t believe the world can achieve net zero emission by 2060 unless global nuclear energy output doubles and reaches 15% of electrical generation.
That is easy for the nuclear energy to say, but please do not make it easy for the industry to sell. It is mostly likely wrong and self serving.
From the article you quotes, it seems that you somewhat misstated the problem. For a refresher, your comment was:
But according to the article,
Lithium-ion batteries have absolutely dominated new storage construction in recent years. But they rarely can deliver their full power capacity for more than four hours — that’s what people mean when they say “discharge duration.” Batteries technically can go for longer, but it generally costs more than it’s worth in today’s market dynamics.
Its not that they lose their charge in 3-4 hours, it’s that they can only deliver their full power for 3-4 hours. So if you want power for 6-8 hours, you need twice as many batteries.
But if you want the power tomorrow, it will still be there.
The article claims that twice as many batteries isn’t worth the costs, but doesn’t really get into the cost problem. I can’t see how any of the hardware is the issue, unless you are bumping up against supply problems. Twice as many lithium-ion cells should cost about twice as much. Twice the supporting hardware (wires and controls and power conversion bits and pieces) should cost twice as much. Twice as much land and buildings should cost twice as much. So I don’t know what the cost problem is. It would have been nice if they got into that a bit rather than just stating it as fact and then dropping it.
Well 30+ year projections are often inaccurate. But China in 2008 predicted 40GWe in 2020 and achieved 50 GWe. So their 20 year projections are pretty good. According to a recent study from Tsinghua University believed to be influential in China’s 5-year plan, a low-carbon future for China requires 300 GWe from nuclear by 2050, a 6-fold increase from 2020.
Low carbon will not happen without significant nuclear power. That’s just the reality of it all. And no amount of anti-nuke ideology will change that.
Okay, but that’s not what the paper you linked to said. You can’t cite the paper when it suits you and then reject it when it doesn’t.
You are probably right. I was writing on memory and that is often a mistake. Here is a reference though that suggests long-term storage is an issue, but not as acute as I made it out to be:
“But lithium-ion’s economics and physical properties limit its storage duration to eight hours of discharge. “If you take a lithium-ion system, charge it, and leave it for three months, it will self-discharge,” says Vincent Sprenkle, technical group manager for the Electrochemical Materials and Systems Group at Pacific Northwest National Laboratory (PNNL).” Cookie Absent
I still find the terminology confusing, but my mistake.
Well, to be fair if you go to my original post, I cited the paper solely for the conclusion that “it would be far more expensive to reach a low emissions world without nuclear energy than it would be with it”, which you will find in noted the paper with respect to Fig 1.5a-e. (I’m afraid this will force you to read beyond the executive summary).
And I don’t reject the paper. I have a different view about dispatchability perhaps but then I never used the paper to support my view. Perhaps you are taking this discussion too seriously?
Frankly, that’s a bunch of hooey. It all depends on the discharge rate. There’s a maximum rate - how fast they can deliver their power without catching on fire. (That’s probably the 3-4 hours mentioned earlier.) There’s a maximum life rate - the fastest you should discharge to get the longest life out of the cell. But there’s no real minimum rate.
I have a few of those little “lipstick” phone chargers that I occasionally use to charge my phone overnight while also using the phone to monitor my sleep. They will run just fine for 8 hours, taking my phone from 60% to 100%, then keeping it there, and still have 2/3 of their charge available. And this is the same technology - the same lithium-ion cell - that is talked about for electric storage for home scale to grid scale.
And it’s also important to note how they would be used in real life. For grid storage, you’re not going to charge them in September and then let them sit until January when you use the stored power. They’re going to be discharged and recharged on a daily cycle. The sizing constraint would be to make sure they are large enough to keep their state of charge from falling below 20% (or thereabouts) at the lowest drawdown point in an annual cycle. In the summer, perhaps the battery cycles between 100% and 30% most of the time, while in the winter they cycle between 80% and 20%. Something like that. Or maybe the numbers are reversed - I’m not sure of what the annual cycles look like. But I’m pretty sure they exist.
The time shifting in power use is in the average state of charge of the battery. If it’s closer to fully charged on average in one part of the year and closer to discharged in another part of the year, you are shifting that energy from one part of the year (or month or week) to another.
The losses over time are quite real - there is some loss of energy over time. But those losses get replaced every day when the battery is recharged. You’d need to account for them in your annual electricity needs (along with other losses on the grid), but they’d be irrelevant on a daily basis.
Again, back to real world stuff, my little phone charging cells often sit for a couple of months unused. I rarely see them lose more than 1/3 of their charge during that time. So those losses aren’t huge.
But, as we all know, you can’t just put new hydro anywhere you want. Same for pumped hydro storage. Same for geothermal. All these are great and should be fully utilized. But they aren’t going to work everywhere.
Peter you are right.
The cost problem is, basically, this: Each time you charge and discharge the grid battery it is worth some dollar amount. If you charge fully and discharge fully once a day (i.e. the typical ~4 hours) you get about 365 of these cycles per year. If you charge then discharge more slowly, over a week, for example you only get 52 cycles per year. This means the battery is costing you 7x as much to own and operate (minus some non-energy operating costs).
If you are time shifting from summer charging to winter discharging, you are only getting about 1/365th the value out of your battery. Of course, this isn’t the whole story since batteries have some limit on total cycles. But they carefully operate them to not use the full deep charge/discharge but to cycle only within the sweet spot to maximize total MWh over many cycles.
Apologies. I thought that when you cited the paper to support your viewpoint, you actually were using the paper to support your viewpoint.
But your story falls apart because 300 GWe of nuclear is only about 8% of China’s electrical generation in 2050, but renewables make up almost 80% of China’s electrical generation of about 3000 GWe or 10 times as much as nuclear in 2050.
And if you look at their 2 °C scenario, then nuclear makes up only about 6% of China electrical generation in 2050, but renewables make up about 88% of China’s electrical generation.
As you note, batteries typically have a life span measured in charge/discharge cycles.
So if you have one full charge/discharge cycle a week vs one a day, the batteries will (in theory) last 7 times longer. So it’s doesn’t cost 7x as much. It costs the same per discharged kWH (plus a bit for the fixed costs, as you mention).
Without deflationary energy policies we actually can not afford the energy we need for most of the population on the planet and we burn out the planet’s environment. Nuclear energy is not a deflationary energy policy. Meaning spending on nuclear energy is a misallocation of resources…but we have known that since the 1970s.
Lithium batteries may be best available utility stationary battery storage for now. But other technologies such as iron batteries or flow through liquid batteries may offer potential for the future.
The EV market needs lithium batteries for their high energy per wt. That will keep the cost of lithium high. Other technologies may be more cost effective where energy per wt is less a concern.
I’d bet on “sand batteries”, which unlike chemical electrical storage using exotic and difficult to refine materials, never wears out and is available in any quantity required at the scoop of a shovel in almost every country in the world.
You can store heat up to 1,000 degrees f (or more, sometimes) for days, weeks, even months at a time; you can make the heat whenever you have a surplus of power (say, when your base load is more than needed, like at 3am), and you can pull it back and turn it into electricity on command with currently available technology using generating turbines or thermoelectric generators (small ones available now on Amazon or Alibaba.)
It’s basically a pile of sand, enclosed in a metal housing, with pipes running through to transfer the heat either in or out, depending on which way you want the energy to go. It’s not exotic, it doesn’t require a lot of R&D, nor is it dangerous in the sense that a nuclear, natural gas, or other hydrocarbon plant is.
The only moving parts are the pumps, commonly available and easily replaceable, and/or the turbines to convert heat to electricity. The technology is amendable to distributed storage, like in your backyard, or on an industrial scale put wherever there’s room for a bunch of two story silos filled with sand.
Silicon era, indeed.
All true. But then you have real estate costs, operating costs and the cost of capital. With one cycle per week it will take you 7x as long to hit break even of the install cost of batteries and time to get your ROI. It would be like owning a car and only driving it on Mondays. Then you have to own another car for Tuesdays, etc. Sure, after 7 years each one only has the mileage of a one year old car.
Tell me this makes sense to the company accountants.
I think that’s the difference of opinion here.
It appears you are assuming that the best use for batteries is to fill them up every day, then exhaust them every day. I’m not making that assumption. To the contrary, I suspect you don’t want to do that.
If you run the batteries to exhaustion daily, you have to turn to some other source of generation once the batteries are done. And if you have that source, why don’t you use it instead of the batteries? The batteries make no sense at all.
I’m talking about sizing the batteries large enough that you DON’T need to run them to exhaustion daily. Instead, they’re large enough to handle the expected peaks and valleys in demand throughout the year. And that let’s you dispense with some other generation plant.
As I contemplate this a bit more, what I’m talking about is the longer term end goal of batteries. You may be thinking about the shorter term - as batteries are coming on line.
In the short term, yes, you probably do run the batteries to exhaustion daily. That does make the most sense while adding to the grid. Those batteries allow some time shifting during the day, which can keep you from needing to add some other kind of generation.
As you add more batteries to the grid, you add to the ability to time shift the generation. The more batteries you have, the longer the time period you can shift, until you get to the point where you can time shift for the whole year.
PS - Feel free to substitute “energy storage” for batteries everywhere you see it above. The key is not batteries, it is what they represent - the ability to generate electricity at one point in time and use it at another.
Perhaps, but I think when engineers who actually work on utility-scale battery storage say stuff like “If you take a lithium-ion system, charge it, and leave it for three months, it will self-discharge,” they do so from the context of real world experience. It is well known for example that self-discharge rates significantly increase at nonoptimal temperatures. For example, one study shows that “Li-ion self-discharges about 5 percent in the first 24 hours and then loses 1–2 percent per month; the protection circuit adds another 3 percent per month.” That’s not too bad, but at at 25C (77F), the energy loss grows 20%/month for a fully charged battery.
The fact remains that most analyses seem to agree that current battery technology is not capable of the kind of long-term utility scale storage you are proposing, at least not in a way that makes economic sense. Hopefully that level of technology will develop in time to save the planet. Maybe. But I think climate change is a sufficiently significant problem that a variety of technological strategies have to be tried in the hopes that enough reach their potential to mitigate the worst of carbon-induced warming. I think nuclear should be part of that effort.
I actually don’t think the utility would cycle them fully every single day, but rather use them strategically based on amount of excess renewable generation early in the day and usage demand later in the day. Each full cycle of a battery has an approximate known cost and the utility operator would need to determine the value based on each day’s projected demand each hour and known generation sources throughout the day.
I don’t think this is the usage model. They charge up early in the day hopefully with excess renewable generation. However if projected demand later is higher than the excess they might charge with other available sources. Then they discharge using the battery as one of many sources to fulfill the peak demand in the late afternoon/evening. There isn’t another source that could be used. And no other source is needed after the peak.
See the duck curve for a visual of how this typically works
Yes I am because the current cost for batteries is (relatively) high and they only really make economical sense for a few use cases – some only time-shift electrical demand for a few seconds or minutes and others (that we are discussing) a few hours, but maybe even a day.
IMO, maybe when they are 5x - 50x cheaper it will make economical sense to shift from week to week or season to season. (Maybe/probably this involves new technologies, innovative thinking or just higher baseline prices for electricity)
See here for some good info on utility battery pricing near the middle to end of this article