Nuclear's role in a net-zero world
knowablemagazine.org
knowablemagazine.org
It’s a surprisingly easy read too. More like a long blog post than a dry paper. One of the examples in there is nuclear technologies, which is what brings this up for me.
It reminded me of a lot of themes present in my design education, of political downstream effects created by apparently neutral technical decisions, and therefore the inherent responsibility of creators when putting something into the world.
Not just standard economic externalities but consequences for human interaction.
A nice flashback to university, so thanks!
For example, a whip specially-designed for the convenience of some humans to inflict coercive pain on others who can't fight back.
Or for a more sci-fi bent, imagine an array of mass-produced devices which can reduce a planetary ecosystem into decaying goo while simultaneously reporting on buried ore veins. It tells you something about the likely values and activities of those who would create and use it.
Robert Mose is mentionned midway, so I'd point out that the 99 percent invisible podcast is currently doing a book club on "The Power Broker" [0]. A lot of the issues touched here are inherently design issues, and I think the whole podcast would be a feast for anyone interested.
[0] https://99percentinvisible.org/episode/power-broker-01-rober...
It’s always been puzzling to me that so many libertarian types love nuclear, which is a technology that tends to encourage big centralized systems and therefore big centralized powers (for many reasons). Solar and wind are much more amenable to decentralized voluntary fluid modes of social organization.
I guess small modular nukes would be better this way, but they still require a lot of security and life cycle management that tends to encourage large centralized administrative power.
Except, Wind and geothermal are more space efficient than nuclear when you look at the actual land use of current nuclear reactors. In theory it doesn’t need much space so several GW could sit on a few hundred acres. Yet, Wolf Creek Generating Station is only 1.2 GW on 9,818 acres that’s worse than many solar projects.
Which seems to be the general issue with the Nuclear industry. In theory you can say all these wonderful things, but in practice people are looking at decades of results when they are choosing to invest or not.
Nuclear doesn’t need technological breakthroughs from cutting edge research, that just leads to more delays and boondoggles. What it actually needs is efficiently run projects on time and under budget.
https://en.wikipedia.org/wiki/Kashiwazaki-Kariwa_Nuclear_Pow...
Or 6.5 GW on 2,300 acres?
https://en.wikipedia.org/wiki/Bruce_Nuclear_Generating_Stati...
4 GW, 3,000 acres:
https://en.wikipedia.org/wiki/Palo_Verde_Nuclear_Generating_...
Kind of odd that you'd cherry-pick the lowest wattage per acre reactor to make an example out of.
So sure, nuclear sometimes uses less land than solar nobody is seriously doubting it, the issue is it isn’t guaranteed and wind and geothermal are both vastly better from a land use perspective. Those 15 MW wind turbines don’t prevent people from farming beneath them, good luck doing anything that productive with land that close to a nuclear reactor.
Anyway, we probably need both, and we definitely need to stop burning fossil fuels, so let's not disparage one in favor of the other.
I never argued that nuclear always lost to solar, I even suggested it should be able to use less land than your examples.
I am arguing it never beats wind.
Put a turbine in the on a 10,000 acre farm and you’ll still have a 9,999 acre farm. Put a nuclear reactor on that same land and you’ll be doing a lot less farming.
A 2015 report, “Land Requirements for Carbon-Free Technologies,” compared the land area that various types of electricity generation facilities would require to produce the same amount of electricity as a 1,000-megawatt nuclear power plant in a year. The results highlight the exemplary performance reliability of nuclear energy facilities as well as the very high energy density of nuclear fuel.
A nuclear energy facility has a small area footprint, requiring about 1.3 square miles per 1,000 megawatts of installed capacity. This figure is based on the median land area of the 59 nuclear plant sites in the United States. In addition, nuclear energy facilities have an average capacity factor of 90 percent, much higher than intermittent sources like wind and solar.
which also asserts that:
Wind farms require up to 360 times as much land area to produce the same amount of electricity as a nuclear energy facility, a Nuclear Energy Institute analysis has found. Solar photovoltaic (PV) facilities require up to 75 times the land area.
Sure, look at this picture of corn growing between windmills. https://www.gasworld.com/story/wind-turbines-and-corn-fields...
Gee I wonder how much land they are actually occupying. Or wowzers let’s talk about offshore wind farms…
They’re using a definition that makes sense in one context and then applying it in a different context where it’s meaningless. Grid operators dislike base load power, the term is actually referring to a downside that it’s pricey to turn off. Now outside of that context it sure sounds great to pair with unreliable renewable energy. Except the downside still exists and you lose lots of money when you turn it off making it a terrible option when paired with renewable energy.
Much as you yourself have down above where the amount of land owned by a company surrounding a nuclear plant is conflated to be the amount of land required by a nuclear plant.
To be clear, I picture a hybrid energy future and can easily link to pictures of sheep grazing under solar panels - I can also link to pictures of large scale solar farms that directly place solar panels on the ground, and others that elevate the panels but have no dual use of the land.
All in all area used is a bit of a side show.
No, they own/control more land, this is specifically the amount used by the nuclear power plant.
“It occupies 9,818 acres (39.73 km2) of the total 11,800 acres (4,800 ha) controlled by the owner.”
The fenced off area patrolled by armed guards isn’t freely available for other uses. Believe me if they could sell off 5,000 acres to build a subdivision the company would happily pocket that money but no the land really is being used.
is not the same as "it requires" ...
One might ask why a 1977 nuclear plant that produces relatively low power has such a large security perimeter ... that is orthogonal to power production.
In this case walls cost more money than an empty field, both give you time to shoot someone trying to cross them and they went with the cheapest option. Thus the land is in use and cannot be used for a city park or whatnot.
Why is that specific plant so far outside the norm and above the median for others?
Why do other nuclear power generation plants 'require' so much less land?
What makes Wolf so atypical?
Why do you insist on using it as a "typical" case when it clearly isn't?
Tell us more about weasel arguments.
Never said typical it’s literally the second on a list. It’s one of the few locations I’ve seen over 4,000 acres so it’s not typical, though presumably I didn’t randomly find the worst example and most didn’t include the area.
Edit: 12,000 acres + a 7,000-acre reservoir, which eliminates the need for cooling towers though it’s got 2 reactors. https://en.wikipedia.org/wiki/South_Texas_Nuclear_Generating... if you include the water behind the dam I think it’s 19,000 acres generating 2.56 net GW, exclude the water and that goes up but presumably dams also get much more land efficient.
Poking around, you can find various justifications for sites needing more or less land. Palo Verde needs a 85-acre and a 45 acre artificial ponds because of its location in a desert. Anything located near a large body of water gets that for ‘free’ and can also make use of the water as part of its security zone a double win, though not quite as much as a wind farms in the ocean. Add in multiple reactors on one location and you can really jump up the density.
Poking around at the other end there’s much smaller reactors on big sites. China has one 200MW reactor on a site with multiple planned future reactors... Which is the thing, the real world is messy you can’t just think in terms of typical or hypothetical but what’s actually used. And of course each of these sites generated zero power for years under construction and will also generate zero power for years or even decades while being decommissioned/decontaminated.
Turbines limit how close other turbines can be located, but nothing stops you using that land for other purposes. Nobody is going to complain their corn, cows, or house aren’t seeing high wind speeds.
Meanwhile the plant operator would strongly object to someone building their house inside the exclusion zone, that land is actually occupied.
The Wolf Creek station come in at 1.02 acres/GWh/year (based on 90% capacity factor)
1) current state of LCOE costs of different power generation methods
2) future LCOE evolution/drop in the relevant power generation methods
"Without nuclear, advocates say, we would need to build far more wind and solar power plants to ensure reliable supplies, doubling or tripling costs over power networks that include nuclear."
Look, if you look at last years Lazard's LCOE report, incumbent nuclear is 5x to 6x as expensive as wind and solar. So, uh, yeah, its cheaper to overproduce capacity than build nuclear. That's from point 1: the CURRENT state of LCOE.
For "new nuclear" to enter the equation in current regulatory politics and costs, it won't power up any new plants for about 10 years.
The second point about LCOE comes into play: solar/wind are STILL DROPPING in cost, and almost more significantly, grid storage will probably see the sodium-ion battery and perhaps the sodium-sulfur battery in that time, dropping storage costs by at least 50%, perhaps 80-90%.
Solar cells will likely see some multijunction silicon-perovskite cell come into production, and possibly a full perovskite cell, in addition to increased economies of scale. I would predict in 10 years the cost of a solar cell is effectively 1/3 to 1/5 of current day prices (inflation adjusted yada yada).
Wind will likely be less revolutionary, but I suspect it will drop still another 50-70% in 10 years.
Meanwhile, this is for nuke plants 10 years out, that really don't have new gen designs ready to go. I will be kind and say plants we started funding now are only 4x as expensive as current solar/wind. But by the time they get built, they will probably be 10x as expensive.
I'm sorry, I love nuclear power, it is so cool. But the nuclear industry needed to solve its problems about 20-40 years ago. It's leadership could have embraced global warming decades ago, but the leadership of nuclear is staunchly conservative, owing to a seething resentment of regulation and government oversight.
I believe there exists, possibly a LFTR, a scalable, cheap, safe, near-zero waste reactor design that can be deployed as industrial power generation and perhaps competitive grid power, when combined with regulatory reforms in the goverment.
But that design and reforms needed to happen about 25 years ago.
Now nuclear needs to hold onto its existing plants, research new reactors (US National Labs have started new nuclear research including LFTRs, again, about 25 years too late), and prepare for when solar/wind stabilize in costs such that a price can be targeted with a true new generation design.
As for grid leveling, old gen nuclear doesn't switch on or off fast enough (LFTR can, maybe pebble bed can too), but it could do "three days of storms" generation. But it still isn't as useful as say natural gas turbine for that (which obviously releases carbon).
Alas the LFTR/MSR design is still unproven for a long term industrial standpoint, and has numerous materials issues. China has an MSR up supposedly for a year or two.
It is true that regulatory barriers, NIMBY, and other politics increase costs and delay projects in nuclear, but this is also a convenient boogeyman for them to point to. If only the "magic regulatory framework" existed to let nuclear projects freely be built then... costs would still be higher than everything but coal.
The LWR just isn't competitive, and I suspect any non-breeding design that can't also transmute waste to usable fuel (LFTR can supposedly do this, I think some other breeder reactors can as well). Because waste solid fuel is either expensive to transport and "dispose" at Yucca and is a whole entire extra bunch of NIMBYism and regulatory oversight, or difficult to reform/reprocess into new fuel.
LWR only works on a small portion of Uranium ore, and even when the rods are formed from that fissile Uranium, only a fraction of the fuel is used until the transmuted byproducts of fission spoil the rod.
LFTR has on-line reprocessing of its liquid fuel for extraction of the fission byproducts, so there is ... kind of ... 100% fuel use, and ... kind of ... no waste to transport. I don't want to handwave away (because I don't know the full extent of) all the chemistry and processing needed to handle that (and there are holding tanks for waste that needs to "cool" over months or years).
Other things I would dispute: solar land use is not a big problem. For one, there is this thing called roofs, and there are a lot of them. There are also lots of parking lots, roads, etc that no one wrings their hands over.
Rooftop solar is a lot more expensive than grid solar (and I think it is a scam that it is that much more expensive) but... it would STILL be cheaper than nuclear to put rooftop solar on houses and equip them with their own storage than build new nuclear plants. Such a program would vastly increase civil resilience to disasters and reduce the need for as much "smart grid" development.
With the failure of NuScale, which all of a sudden is the people "not doing it right" for the industry, the actual viability of next gen nuclear took a big hit, and with it basically the last hope of nuclear being a part of the next 20 years of power generation outside of "keeping the current ones on". If the nuclear industry had a usable SMR design about 10 years ago specifically for industrial application, nuclear would be in a healthy place.
And of course, nuclear needs to compete with the ultimately scalable power source: solar and to a lesser extent wind.
Now, I am no Nuclear Engineering PhD-wielding expert, but I've been professing this opinion for a while on here, mostly in hopes of a constructive counterargument or hope for "new nuclear", but ... basically crickets. Which leads me to conclude that nuclear is just shilling these puff pieces to desperately keep the lights on or keep governments investing in whatever new projects there are. And to emphasize, governments should keep up heavy research into new designs for nuclear: there are simply far too many powers-of-ten in energy density involved here.
[1] https://www.un.org/en/climatechange/net-zero-coalition
[2] https://climate.ec.europa.eu/eu-action/climate-strategies-ta...
[3] https://www.sustainability.gov/federalsustainabilityplan/emi...