It appears that the LCOE for utility-scale solar which you're referencing here, community and rooftop both being significantly more expensive, is based on existing installations. But existing installations would have used the cheapest available land, and the cheapest available land tends to be in short supply. Solar installations on a significantly larger scale would have to use less ideal land (less accessible, more rugged, more expensive) and would incur associated cost increases. Solar is like a huge orange tree where we have been mostly picking from the lower branches — there are more than enough oranges on the tree for everyone, but you can't expect them all to be as easy as the low-hanging fruit.

However, I am optimistic about storage, particularly since zinc-bromine seems poised to break into the market, with excellent resource availability. Zinc production is about 13 Mt/yr [1], and the battery offers about 67 Wh/kg, with ~1/3 the weight in zinc, so 200 Wh/(kg Zn), so potential production is over 1 TWh/year before running into availability problems. There are also about half a billion tonnes of bromine in the Dead Sea alone [2]. (Since this is my third Zn-Br post, I'll add that I don't currently have investments in them, but I'm considering it.)

1: https://en.wikipedia.org/wiki/Zinc#Production

2: https://en.wikipedia.org/wiki/Bromine#Occurrence_and_product...

Cheap land is cheap specifically because there is a very great deal of it. But it is a stupid place to site solar, anyway: there are much better places, already in use, that may continue in that use, but with added solar revenue.

The overwhelming bulk of utility scale storage will not be in batteries. They cost too much per kWh stored. Bulk storage will be in media where incremental kWh are cheapest. Think tankage.

>The overwhelming bulk of utility scale storage will not be in batteries. They cost too much per kWh stored. Bulk storage will be in media where incremental kWh are cheapest.

One side of this comparison has cost estimates, the other one is vaporware. You need a certain generation capacity for hydrogen (or whatever) measured in watts, not watt-hours; these plants require maintenance and operation costs likely much higher than batteries. With battery costs approaching $100/kWh, many TWh of storage are attainable.

I'm assuming that nuclear will be a significant part of the energy supply, so I would not expect it to be necessary to, for example, store six months' worth of energy for the winter.

Nukes will soon be unable to find buyers for their power at a price they can offer, most of the time, so will be unable to collect enough revenue to continue operating, anywhere they are not propped up from money coerced by taxation.

Cheap storage will displace them.

There is an enormous amount of cheap land, globally. Any land that is cheap enough for farming is cheap enough for PV, and land that isn't even adequate for farming tends to be even cheaper than that.

Apparently, my use of the phrase "cheap land" was misleading. The acquisition price is one thing. But then you have site suitability, regional suitability, site access (running powerlines/roads), and construction labor availability. Any of these can drive up costs. Desert sands are soft and blow around, rocky hillsides are difficult to maneuver equipment on, forests have to be cleared, et cetera.

Nuclear uses far less resources and land area than both wind and solar plus storage. So sure, it costs some more money but it costs less actual real stuff.

Furthermore, it is not feasible to power individual personal vehicles or homes with nuclear reactors, so using nuclear for the grid frees up those resources that can be used for other stuff, for that stuff.

You do know that you can exchange money for “actual real stuff”, including natural resources and land? So that why we cite costs in “money”.

We cite costs in money because it’s useful. But you can’t eat money and you can’t build from it. You need actual stuff and land to do so. That is limited in a way that the money supply isn’t.

Case in point, the currently ongoing global supply shortage.

I would think the point is that if nuclear costs more money, it's suggestive that it doesn't really use less resources than other options that cost less. Of course, that's not entirely true, as some of the costs for nuclear are insurance, which isn't priced particularly rationally, and of course, for human expertise, but that's arguably a far more scarce resource as are land and raw building materials - i.e. if you're using highly-skilled human brain power to design/build/run nuclear power plants, it prevents us from using that brain power to do other potentially more productive things (which might include coming up with better solutions for our environmental challenges). To be clear - I'm quite supportive of more nuclear power, but certainly not because it uses less land or resources.

I realise that, but this argument is based on the assumption that things in the world are rationally priced.

When an NFT sells for more than a small power plant and a celebrity dog walker can be paid more than a nuclear engineer I don’t have a lot of faith in that premise.

Fair call. But while obviously imperfect, cost in $ is about the best estimate we have of the underlying value of all resources involved for a given economic output. Unfortunately it typically doesn't include externalities - if it had done so historically nuclear might already be cheaper now than fossil fuels.

Uh? Try reading my comment again. That the federal reserve in principle could print unlimited money is irrelevant here.

Land use is a substantial concern for renewables and energy storage. We pay for electricity in marginal dollars. Currently we are building solar in low cost/utilization land. At some point prior to full rollout of renewables, this “cheap” land will run out.

Land use is of absolutely no concern for renewables or storage.

Solar and wind coexist synergetically with existing uses, so renewables do not need any dedicated land at all. There is enough existing reservoir area to float solar on, alone, for many times the world's conceivable power needs.

Solar farms are built in deserts only because idiot investors want them there. They are fantastically less efficient there, and degrade in the heat.

Concern trolling about land demand is the low road.

Why do you believe that land use is absolutely no concern? What data do you have for this?

https://www.nature.com/articles/s41598-021-82042-5

Notes that at 80% penetration, solar will occupy roughly 5% of total land. The paper goes on to study methods of ensuring that this land use change does not drive net carbon increases from vegetation loss etc.

You didn't read or understand the posting you replied to. Why should anybody expect you to read or understand anything else?

I’d suggest reviewing the site guidelines on comments.

https://news.ycombinator.com/newsguidelines.html

I posted a response on some of the challenges with marginal economics in renewables and responded to your comment with a peer reviewed and cited journal article on these issues and the relevant data point from this study.

Your response contains an ad-hominem personal attack, and does not contribute to the discussion.

Which part of a reservoir surface do you consider "land"?

Presuming a total adoption of solar across all reservoirs on the planet. We would have roughly 220k sq km of available “land”, roughly enough to support the EUs energy demand (assuming the energy could be magically transported from reservoirs on other continents)

Placing panels everywhere we can is a great idea, however pretending that there is a silver bullet is putting our heads in the sand.

Solar is not the only clean, cheap energy source. Wind coexists nicely with agriculture.

And, constructed reservoirs are not the only still water. California has estimated it can gather 12 GW just from its canals.

Solar may also usefully be sited on active pastureland, which is not in short supply. There, it reduces water demand and offers shelter for livestock. Livestock keeps down weeds.

There is a fair amount of industrial roofing in use, where solar extends its life. Some people have already installed there.

> Solar is not the only clean, cheap energy source. Wind coexists nicely with agriculture.

Then why not nuclear too? The world sorely needs options where we can spend some CapEx and solve global warming. I see nothing wrong with a mix of hydro, solar, geothermal, wind, and nuclear. There is no reason not to expect capital efficiencies in larger nuclear build-outs compared to the sporadic reactors we've built for the last ~50 years.

Right now we are seeing coal and natural gas plants getting built because there aren't enough renewable options. As has been the case for the last 50 years while renewables catch up.

Simply, capex $1 spent on renewables gets you several times what you would get buying nukes, and now, not in 2035.

Globally, there is no shortage of cheap land for a renewable-powered economy. If locally land is expensive, don't put energy-hungry industry there.

This may mean Europe ends up deindustrializing vs. sun-soaked places closer to the equator. I suppose King Leopold wouldn't have liked that.

This is a fair statement, major industrial metros such as Chicago emerged due to proximity to resources (iron, grain), transportation (lakes, rivers, rail), and energy (coal).

There surely will be a new equilibrium including newer power generation means, however many sunny places are far from water and other industrial inputs - moving steel mills to New Mexico is not quite a slam dunk in terms of transportation.

The link you provide seems to say that the cost of wholesale PV + storage is $85-$158 (bottom figure, line 3). Am I misreading that?

I’ve noticed that many solar+storage installations these days are 4-hour storage, so not sufficient for baseload. I think the number would be higher if we were shooting for baseload from our storage.

This site is using a benchmark of 4 hours for storage. This is sufficient for California, but not really anywhere else. Here in Finland, it would be literally 3 orders of magnitude too low. This, er, has an impact on cost.

> This is sufficient for California, but not really anywhere else. Here in Finland, it would be literally 3 orders of magnitude too low.

Yes, but compared to California (and all the other very sunny parts of the US and the world where huge populations live), from an energy consumption perspective, tiny Finland doesn't really matter much. Or put another way, it matters far more to the future of the climate to decarbonize energy production in California than Finland.

That said, companies in Finland have been developing some neat grid scale heat storage batteries lately, so there is an opportunity for them to have a big technological impact that way.

Yep. SMR is a much better drop-in replacement for fossil sources than renewables (generally stable/controllable output, dense locating so no expansion/upgrading of grid required unlike solar and wind, and can provide large winter outputs in more northerly climes where solar efficiency fades in the winter).

Wait, am I reading it correctly that your reference is saying that the cost of fuel and operation of a nuclear reactor alone are already around the same value as the total costs for PV solar?

I didn't expect that.

Just the turbine part of any power plant that converts heat to electricity is expensive enough that it has problems competing with PV in sunny locations (if you only include a little bit of storage in your calculations)

What I got is that if you had a perfectly fine already working nuclear plant, completely paid off, it's about as profitable to abandon it and build a solar plant as it is to run it.

It's the same as the most expensive residential rooftop solar. Nuclear is about 8-10x the total cost vs utility scale solar projects.

There's no viable way to build storage at the scale required to run a wind and solar grid. Even building just 1 hour of electricity storage amounts to 2,500 GWh. The entire world's annual output of battery storage is somewhere between 300 and 400 GWh. Any attempt at grid scale storage would lead to shortages driving up prices.

Similar bottlenecks occur with pumped hydroelectricity. To build it economically you not only need an alpine lake handy, it also needs to be close to transportation infrastructure. As those sites are developed, we'd turn to more and more remote sites.

As has been pointed out to you many times before, there is no need to use batteries to store utility-scale energy, so battery production capacity is immaterial. Mentioning battery production of the moment again is disingenuous, not to say dishonest. Production capacity, anyway, increases with demand.

The key principle of bulk energy storage is E = Fx, applied liberally worldwide for centuries. It is taught to every freshman engineering student. Nothing blocks further application of the principle, at any scale.

And as has been pointed out numerous times, no existing alpine lake is needed for hydro storage. And, the site does not need substantial "transportation infrastructure". Dozens of hydro dams, still in use, were built in California's Sierra Nevada mountains in the 1920s via roads a car cannot use.

> The key principle of bulk energy storage is E = Fx, applied liberally worldwide for centuries. It is taught to every freshman engineering student

Hi there, I graduated engineering and am currently applying for professional licensure. I took only one course focused on energy, but I've never heard of E=Fx. Googling it brought up no relevant results. Could you expand on what it is?

They are likely referring to force * distance = energy. Which in context is a pitch for pumped hydro storage.

This all sounds well and good, however it’s highly unclear that utility scale pumped hydro is viable. To make it work you need to have plentiful water, limited evaporative loss/other losses, and two large basins to store both the charge, and discharge of water.

In the event of drought, these facilities could become impractical. Hydro facilities have their own environmental concerns. Combined with shifting climates and rainfall patterns there are many challenges to be solved. (some of which go away if the hydro storage is in underground ceiled chambers… which also has a cost associated)

Pumped hydro is not the only storage; otherwise I would have said E = mgh.

Pumped hydro is practical in many, many more places than have hydro generation today, because unlike those, it does not need a watershed. It can use a deep underground cavity where a hill is not forthcoming. Where water is scarce, other methods will be used.

Other applications of Fx include (but are not limited to) compressed air, and buoyancy.

Places that run low on storage resource can import and burn fuel, as they do now, or schedule power from a transmission line, where they have one. Soon, liquified anhydrous ammonia will be cheapest and most practical, but liquified hydrogen may be cheaper and sufficiently practical for bigger utilities. Ammonia has the advantage that it does not need cryogenic treatment. These will be available from numerous tropical sources.

You may go back to your freshman physics textbook.

If you get a professional engineering license without recalling basic Newtonian physics, that just tells us the licensing process has utterly failed us.

> You may go back to your freshman physics textbook.

I may, but I live in a country that doesn't use the word freshman, so I don't know which year physics text that would be.

I assume from context you were meaning to refer to gravitational potential energy, which is generally represented as ∆U=mg∆h (change in gravitational potential energy equals mass times gravity times change in height). The variables you used (E=Fx) would mean energy equals force times (variable). From dimensional analysis the variable would have to be units of length (m), however it clearly only applies in the vertical dimension so it's better to use a specific variable like ∆h

Regardless, energy storage via gravitational potential energy (i.e. pumped storage) has been in use for about a century, and still makes up a rsmall proportion of grid operations. It has very specific requirements that aren't available everywhere. It's not an end-all solution for energy storage.

Not the parent, but feel compelled to clarify the physics here. Delta E= force * distance always holds. This is the same physics of a car breaking or a jet accelerating.

However a more general formula would be the integral of net force over distance.

Yes, I was just suggesting that force times distance is not relevant to energy storage as gravitational potential unless the distance is vertical.

Gravitation is not the only force.

Even without local "freshman" classes, you may look it up in any dictionary. Or, just stop pretending.

I understand gravitation isn't the only force... how are you proposing to store energy? Other options (beside pumped hydro as gravitational potential energy storage) that come to mind are flywheels, springs, or batteries .... all of these technologies have existed for more than 100 years, and batteries are the only ones currently growing in usage - there's a good reason for that.

Batteries are being used primarily in electric vehicles, for, as you say, good reasons, and for domestic backup. For bulk utility storage, batteries are the most expensive choice, although battery cost is still falling fast and numerous utility-adapted chemistries are competing.

F may be water pressure, as in pumped hydro (which is growing) using elevated or underground reservoirs, or air pressure, as in CAES underground or underwater compressed air, or buoyancy using sea-floor pulleys and floats. No doubt as a soon-to-be Licensed Professional Engineer you will soon be able to think of other persistent forces.

Springs and flywheels will not be used for bulk utility-scale storage.

Generally, utilities will use what is cheap and reliable at the time they build it. Building storage before you have enough spare renewable capacity to charge it would be a bad misallocation of capital

Anhydrous ammonia will not be the cheapest medium, but has advantages of transportability and fantastic usefulness. Any unused overbuilt capacity will be put to work synthesizing for sale.

What do the values of E, F, and x even stand for? Googling this yields no related results [1]. Presumably F and x stand for the energy output of a storage system and the duration that system is online, respectively. But simply stating that equation is pointless, unless you actually give a plan to deliver on appropriate values of F and x. It's about as inane as saying "E + MC^2 so let's just build fusion and be done with it."

1. https://google.gprivate.com/search.php?search?q=E%20=%20Fx%2...

You really are fooling no one. Why maintain the imposture? Are you not embarrassed at being called out on falsehoods each time you trot them out again?

For the third time, what do F, E, and x stand for? Are you not embarrassed about telling people to go back to their physics textbook when the equation you're writing has nothing to do with energy storage?

You still fool no one.

Then what will be used for grid-scale storage if not batteries and hydroelectricity?

> And as has been pointed out numerous times, no existing alpine lake is needed for hydro storage. And, the site does not need substantial "transportation infrastructure". Dozens of hydro dams, still in use, were built in California's Sierra Nevada mountains in the 1920s via roads a car cannot use.

Such as? How did they get heavy machinery to these dams to build them? All of the dams I can find like the Shasta dam, Orville dam, etc are in fact close to major transportation infrastructure.

The Sabatier process has significant blockers to actually being used: First, it needs a source of carbon dioxide, and the small concentration of it in he atmosphere is insufficient. Biofuels could be used for this, but they aren't produced in sufficient quantities. Second, it needs a source of hydrogen. Almost all hydrogen produced today is done through steam reformation, which emits carbon dioxide. Electrolysis accounts for a slim minority of hydrogen produced today, mainly due to inefficiency and issues with electrodes corroding. Nobody is operating commercial power to gas storage facility, it's only prototypes.

Commercial power-to-gas facilities are in operation in Germany, but large scale investments make little sense right now. The money is better spend in increasing the supply of renewable energy, or electrifying things that currently burn fossil fuels. Days where we have too much renewable energy are very rare still. It would be a waste to turn it into Hydrogen.

> Commercial power-to-gas facilities are in operation in Germany

Source? I'm only aware of pilot programs in Japan and Germany, not commercial operators.

Interesting. Though the lack of any cost estimate for this 1 MW facility does not inspire confidence.

Yes, and in 1950, no nuclear power plants operating; therefore nuclear power is impossible.

Who do you imagine you are fooling?

Disingenuity is deceptive everywhere it appears. Know that you are not fooling anybody.

I agree: nobody is fooled by your baseless claims that we can build dams where there's no transportation infrastructure. How does heavy machinery get to the site? How does the concrete get poured?

You insist the dams high in the Sierra Nevada range of California don't exist? You choose a strange hill to die on: anybody can see them in Google Maps satellite view.

They are not generally made of concrete, not being deep enough to need it. Their high pressure is confined in the penstock well downslope.

Again, just name three that aren't accessible by road or rail. If they're so easy to spot on Google maps, it should be trivial for you to find them.

Nice try. But I will not be wasting more time on your behalf.

Earlier you claimed it was trivial to find these inaccessible dams. You took the time to respond to my comment, and could have easily taken the trivial step of finding a few of these dams but curiously chose not to. Or, maybe these dams don't exist.

No such inaccessible dams exist, and if they do they hold an inconsequential amount of hydroelectric energy. Just exercise a modest amount of reasoning: no transportation infrastructure means all work is done by hand and all material transported manually as well.

Pre industrial dams did exist, but they were very shallow and used for irrigation rather than power. Moving enough earth and concrete to make a usable gradient is not possible without heavy machinery.

Pointing to one or two current storage technologies as evidence that the problem is fundamentally unsolvable is quite a logical leap.

Can we apply the same reasoning and conclude that because existing nuclear reactors are inadequate, no possible future nuclear reactors can be adequate?

Why does pumped hydro have to be close to transportation infrastructure, btw? Can one not build roads or rail lines?

The point is that no existing storage system is sufficient and thus solar and wind is really solar, wind, and fossil fuels. And nuclear power is indeed sufficient. They can indeed modulate power output, and we have the capacity to build them at the required scale. Some countries already have, like France.

> Why does pumped hydro have to be close to transportation infrastructure, btw? Can one not build roads or rail lines?

You can, but the more remote the build site the more expensive it becomes.

Current nuclear technology isn't sufficient either. A world powered by nuclear burner reactors runs out of uranium very quickly. Breeder reactors or seawater uranium extraction would be needed, and neither of these are "existing" in the sense you're using for storage.

This is incorrect, even existing supplies with no different form of extraction would last over 200 years: https://www.scientificamerican.com/article/how-long-will-glo...

Nuclear seawater extraction is already conducted, at a price of about $200 per pound [1]. Raw fuel extraction is a small portion of nuclear's cost (enrichment is a much bigger portion) so this is not a significant cost increase.

1. https://www.forbes.com/sites/jamesconca/2016/07/01/uranium-s...

200 years at today's consumption rate.

To power the world with nuclear would require many more reactors. Current world average primary energy consumption is 18 TW, which is 6000 3 GWth reactors. That's an order of magnitude more than currently are in operation. Add in demand growth from industrializing countries and the uranium would last maybe a decade.

Uranium availability is not an immediate problem if you assume nuclear remains a sideshow on the global energy stage. That's to assume nuclear is neither needed nor particularly useful for addressing global warming. Is that what you're assuming?

Did you read the article?

> However, seawater concentrations of uranium are controlled by steady-state, or pseudo-equilibrium, chemical reactions between waters and rocks on the Earth, both in the ocean and on land. And those rocks contain 100 trillion tons of uranium. So whenever uranium is extracted from seawater, more is leached from rocks to replace it, to the same concentration. It is impossible for humans to extract enough uranium over the next billion years to lower the overall seawater concentrations of uranium, even if nuclear provided 100% of our energy and our species lasted a billion years

Yes, I did read it. Did you lose track of what I said? I'll help by quoting myself:

"Current nuclear technology isn't sufficient either. A world powered by nuclear burner reactors runs out of uranium very quickly. Breeder reactors or seawater uranium extraction would be needed, and neither of these are "existing" in the sense you're using for storage" (emphasis added)

So, your comment about seawater there doesn't contradict what I wrote at all.

Again, from the same article:

> And these advances by PNNL and ORNL have reduced the cost by a factor of four in just five years. But it’s still over $200/lb of U3O8, twice as much as it needs to be to replace mining uranium ore.

> Fortunately, the cost of uranium is a small percentage of the cost of nuclear fuel, which is itself a small percentage of the cost of nuclear power. Over the last twenty years, uranium spot prices have varied between $10 and $120/lb of U3O8, mainly from changes in the availability of weapons-grade uranium to blend down to make reactor fuel.

> So as the cost of extracting U from seawater falls to below $100/lb, it will become a commercially viable alternative to mining new uranium ore. But even at $200/lb of U3O8, it doesn’t add more than a small fraction of a cent per kWh to the cost of nuclear power.

This is technology that actually has a demonstrated cost. Moreover it doesn't need to get cheaper at scale since raw extraction is such a small portion of nuclear power's cost. It's not like synthetic methane or hydrogen storage where it's all white papers promising cheap cost, but not actually delivering any storage systems at that cost.

Yes, and none of that is demonstrated, in the sense you seem to demand for renewables and storage. Lab exercises are smaller than what would happen in the field by many orders of magnitude and are happening in idealized conditions. The cost estimates are aspirational. This hypocrisy of yours is what I've been pointing out.

Extracting uranium from seawater on the scale needed to fuel burner reactors requires massive engineering. The estimate I've seen is that fueling one 1 GWe burner reactor takes a seafloor uranium absorption field (suspended in a strong ocean current; if you have to pump the seawater yourself it's already too expensive) of 170 square kilometers. The power/area is already much lower than the time-averaged output from PV.

Extrapolating from the lab bench to 170 sq. kilometers (x 6000 for the number of reactors needed to power a nuclear world) is a far larger stretch than extrapolating renewables and storage to what they would need. Your selective doubt is not being driven by any honest impulse.

> Yes, and none of that is demonstrated, in the sense you seem to demand for renewables and storage

All I demand is that people deliver at the specified cost. Not, for example, a prototype with unspecified cost or a pilot program with much higher cost to and promises that it'll get cheaper at scale.

> Extrapolating from the lab bench to 170 sq. kilometers (x 6000 for the number of reactors needed to power a nuclear world) is a far larger stretch than extrapolating renewables and storage to what they would need. Your selective doubt is not being driven by any honest impulse.

I could say the same about you. Nobody is seriously considering pumping seawater to filter uranium. This is a totally bad faith argument.

And you give this estimate of 170 without actually giving a source for it nor specifying details like how far apart the buoys are spaced, or if reprocessing is used (which cuts down fuel use by more than an order of magnitude).

The line you’re presumably referencing is for four hours of storage.

That’s nowhere near enough to sensibly compare with a nuclear plant.

These “levelized” costs are always comically inept.

- Assumes solar installation in a sunny climate. It’s an order of magnitude difference in potential solar output between San Diego and NYC.

- Underestimates battery requirements by multiple orders of magnitude. Again, because it has comically inept assumptions.

- We literally cannot deploy solar at scale today. We do not have anywhere near the battery production necessary for utility scale deployment. We would need to double our worldwide battery manufacturing ability, double it again, double it again, and then double it one more time.

- Massive demand for solar and battery, combined with finite production capacity, would lead to dramatic increase in prices.

TLDR the other energy technologies can actually be deployed at estimated prices. Solar cannot be deployed at scale globally. At all. Period.