>Environmentalists are divided over nuclear power, with some maintaining it is dangerous and expensive, while others say that to achieve net zero emissions by 2050 all technologies are needed.
Yes, a mix of technologies. But there's more to it than that. We're going to need increasing amounts of power in the future to explore the solar system and to solve all kinds of existential problems. Energy usage is not inherently evil provided we learn how to do it safely, which is an ongoing process.
> ... with some maintaining it is dangerous and expensive
Expensive, probably. Dangerous? Not according to the data. Lowest deaths per terawatt-hour of any energy source including solar. Yes, even if you count Chernobyl, Fukushima (where 6 folks died) and Three Mile Island (nobody died) [1]
You have to understand that "Lowest deaths per terawatt-hour" is not the sole stat to look at to determine something's danger.
I think the flat amount of direct and indirect death, as well as considering the amount of people who's quality of life has diminished, but not enough to kill them.
Other energy industries have these problems too, so I have no idea how they actually compare.
I think the flat amount of direct and indirect death, as well as considering the amount of people who's quality of life has diminished, but not enough to kill them.
How do you measure that? Why do you think is higher for nuclear than for other technologies?
I've actually wondered if we could find enough suicidal idiots maintain the growing number of wind farms.
Finding climbers for static towers is already hard enough. Death rates of cell tower climbers are 10x that of normal construction workers (cell tower climbers have the highest death of all construction jobs). Of all those deaths investigated by OSHA, almost 40% involved no rules violations -- speaking of that, only 3 of those found in violation were fined more than about $25,000 with most being fined less than 10k and some only being fined a few hundred dollars).
Unlike static cell towers towers, you have lots of moving parts you can't really stop and high-voltage power sources you can't completely shut down. There's additional risks of fires and even disintegrations.
Unlike cell towers, I can't find any overall statistics. The closest I can find is [this organization](http://www.caithnesswindfarms.co.uk/AccidentStatistics.htm). It's crazy to realize that at least one windmill suffers structural failure every month. Two catch fire and (since there's currently no way to put out a fire hundreds of feet in the air) burns down. Two more will suffer blade failures whipping sections of blade out at up to 325km/hr (200mph). There will even be someone who is injured due to ice being flung by the blades and hitting someone over 150 meters away. This is all before counting the couple people killed every month. And to top it all off, they don't even have complete statistics.
Underwriters Lab (the official unofficial US government lab) in 2015 claimed a 0.54% blade failure rate worldwide, but with almost 500K units, that's still thousands of units every year (I'd note that they don't actually have reliable data on the 42% of all wind turbines that happen to be located in China). [source](https://www.enr.com/articles/42352-are-four-wind-turbine-fai...)
> There were 737 reported incidents on UK offshore windfarms in 2016; blades falling off, turbines tipping over, falls from height, vessels sinking in ice-cold water, groundings, onboard fires, helicopter crashes make up just some of the reports. The most common accounts were of hand injuries, while fingers cut off, arms crushed, broken bones, fractures, lifting injuries and teeth knocked out also occur. Non-accidental medical emergencies include strokes, heart and asthma attacks, and anaphylactic shock.
> Of all the incidents at UK offshore windfarms, the majority happened on operational sites: only two were recorded during windfarm development in 2016. Around 44% of offshore medical emergencies occurred in the turbine region, while just over one quarter were on vessels. The number of fall-related injuries was 110 or 15%, of which 95 (13%) were during heavy lifting operations.
Then there's the issue that wind turbines seem to have a realistic lifespan of only 12-15 years instead of the 20-25 years claimed and lose half of their total power output over that 15 years. [source](https://www.telegraph.co.uk/news/earth/energy/windpower/9770...).
Solar panels have a relatively low direct body count, but mining then melting down entire mountains for their rare earth elements has a severe environmental impact (not to mention the environmental toxins and toxic waste produced during the actual manufacturing process).
In the entire existence of US nuclear power, there have only been around 60 incidents which resulted in death or damages over $50,000 (and only 13 deaths overall) and unlike wind power, every little thing about nuclear plants is logged thoroughly. Even the "waste" is safer to store on average than the caustic waste from manufacturing and will be refined and reused once cheap mining sources dry up.
Rather than scaring everyone with nuclear FUD, we need to embrace it as the most promising and safest green energy technology we have.
EDIT: I'd also give a shoutout to concentrated solar which could be a great green daytime alternative with some caveats (variable power output, still needs nuclear power at night, much more geographically limited, etc).
It's not "probably" expensive; it's undeniably very expensive, which is why it's a non-starter in most of the world. Look at South Carolina's fiasco with nuclear...the only country doing nuclear is China, which is known for massive infrastructure spending.
Horror story from a friend who worked in China as a consultant on the power grid back in the mid-aughts. Basically the guy who was leading construction on the plants was working with another who was to supply the concrete for the walls. Problem was, the concrete guy was behind on his "five year plan", so plant guy didn't have enough concrete. Plant guy still has to meet _his_ targets, so he just builds the plants with thinner walls.
Sure it's an anecdote, but the point is China has very loose safety standards and cuts a lot of corners. She's much more okay with losing a few people here and there.
Horror story from my stepfather at Bruce Nuclear Power Plant in Ontario Canada: At least twice a year there was a "secret leak" where he had to come home early but no media was alerted and they had to hire racialized folks from the city to go in and make repairs. He also stole reactor sealer to seal our basement from spring leaks and our basement leaked. And on and on. Canada has very loose safety standards and cuts lots of corners.
> At least twice a year there was a "secret leak" where he had to come home early but no media was alerted...
Uh, [citation needed].
> ... they had to hire racialized folks from the city to go in and make repairs...
Racialized... folks?
> He also stole reactor sealer to seal our basement from spring leaks and our basement leaked.
That might be because, and I'm speculating here, your basement isn't a reactor. I've heard that a sealant for a specific kind of material may not work on literally any other kind of material.
> Canada has very loose safety standards and cuts lots of corners.
While a fair point, no, the last mine stopped operating in 2011 in Thetford Mines, and a total ban was brought in around 2018. The government is investing in some technologies to extract magnesium from the left-overs while neutralizing the remaining fibers [1].
No,France has not built any new nuclear power plant in decades, the only one being built (flammanville 3) is a mess, and there are goals to reduce the share of nuclear to 50%.
No. France is retiring plants and has a nominal plan to retire 30% of their nuclear generating capacity in the next decade.
There is one plant being built at Flammanville. It is 11 years over schedule and 400% over budget, and still not certain that those will be the final numbers.
That's because the international elites that are really running France have decided that it'd be easier to just shift pollution to China and make their money there. There's no sane argument against nuclear today, particularly if you believe in climate change. And yes, you can make anything cost 400% as much, just pile on more regulation and NIMBY-ism, and you'll get there eventually, that's not news.
The Flamanville 3 EPR nuclear reactor is a mess (and I've seen it first hand, working on one of its subsystem a few years ago). The regulations bodies, aka the ASN (Agence de Surete Nucleaire) played its role, and uncovered various issues, most worrying, defects in the reactor vessel itself (and also there was some attempt by the manufacturer to hide these defects). And it's only one of a long chain, the were others like concrete being poured without/insufficient rebar or improper composition, various welding issues on pipes or machinery. And each time, it was not some minor mishaps requiring a quick fix but a major mistake requiring undoing what was done, redoing it properly and causing months of delays.
Basically, this is an Engineering and Project Management failure. On this kind of giant projects, logistic, coordination and management is key, the design is also key (system of systems, interfaces between systems etc) and it failed spectacularly here.
From the political side, despite some calls to just stop the construction from minor parties, the commitment from the politicians to see it built has remained strong all along. And there was also a deep commitment from the French state to the Nuclear industry, with the government bailing out Areva just a few years ago.
The Finish one is not better off, with the same kind of delays. The Chinese ones are in production (after significant delays) but 1) They learn from the mistakes of the first twos 2) They have probably better experience in large projects given the past construction boom in China 3) Some issues were probably put under the carpet as the Chinese government is not exactly renowned for its openness.
The point is that it's an extremely safe method of power generation, and what is cited as a key downside is in fact empirically a huge upside to nuclear. No one is demanding retrofit rooftop solar, wind, or gas turbines be shut down on safety grounds, even though they are considerably less safe.
As far as humans are concerned overall progress has been linked to climbing up fractions of the kardashev scale since fire was invented.
And even before fire, organisms could only become more complex as the energy available to them increased. You can't run a hummingbird or mammalian brains on photosynthesis directly, it's just not energy-dense enough.
This thread is becoming recursive. Is organism complexity progress? Everyone might agree with you, or not. That's because progress is whatever we say it is. It's subjective, in that everyone can have a different view of it, and it's non-constant, in that you can change your mind whenever. Progress is such a loaded word today. I'm hardpressed to find more than a few words with so much cultural baggage. Even someone who agrees that good and evil are in the eyes of the observer, can often say that the definition of Progress is self-evident.
Most quality of life improvements do take energy, and yes, quality of life is subjective, it is conceivable to imagine a mind that takes joy in their child dieing of a preventable disease, some vertebrates eat their young, but overwhelmingly, humans like to see their child get old enough to play and talk as well as laugh and cry.
Things like reliable and easy access to clean water, refrigeration for medicine and food, shelter against the weather, clothing, cleaning mechanisms for the above, all of these are non-equilibrium phenomena and so take energy.
So we either need more sources of power or fewer people. I'm glad that birth rates are falling, but lots of people don't have those basic technological tools, so energy technology will need to be rolled out or our definition of empathy will need to change to allow for lots of people to suffer.
While one would agree with your list of desirable non-equilibrium phenomena, the need for more power or less people is non con-sequitur, as you fail to show how the listed things require more energy than our current world consumption.
AFAIK, more than enough food and clothing for all people is currently produced, shelter and cleaning probably isn't far, refrigeration for medicine and required food wouldn't take much of industrial output and problems with access to clean water is often caused by "progress" and could be solved with more strategy rather than energy.
Either way, I think most of current energy consumption is for things like heating / cooling inefficient homes, manufacturing things people don't really need, inefficient transportation, brain-dead things like making oil from tar sands etc.
I am saying that for many or perhaps even almost all measures of progress we want to apply, i.e. the specific definition doesn't matter, that progress was either enabled by or directly required more energy being available.
Certainly individual conceptions of "progress" will vary, but only the most deranged practitioners of self-loathing would make the argument that gaining the freedom to exit a destiny of brute survival does not constitute advancement.
Progess is a human invention, without us it has no meaning. Complexity is a first order axiom; if you don't adhere to that philosophy then it is not. You are free to move to a remote area and live off the land.
even before fire, organisms could only become more complex as the energy available to them increased
Heck, look at how the complexity of life increased as soon as mitochondria were invented, going from genomes measured in millions of base pairs for bacteria to billions for amoebas once cells were no longer caught in the square/cube trap of respirating over the cell wall.
Right but I'm also weary of the kind of second-guessing about the future and the nature of 'progress' that leads to a kind of epistemological paralysis. Taking an approach of 'it's all very complicated, who knows what the future holds, let's not try to plan for it' is also wrong.
Exactly. Hegel developed a theory that historical progress goes always forward, or towards better things. Capitalist-based theory says that Capitalism is the inevitable outgrowth from Feudalism, which is false (Meiksins-wood showed that Capitalism is an idea that was enacted through force, not an inevitable outgrowth of the previous systems). Communist theory says that Capitalism inevitably fails and tends towards Socialism, which eventually develops into Communism, which has yet to be shown. People tend to think that Evolution results in better and better organisms, rather than those that convey the best advantages for it's environment. For some reason lot of people tend to think that a society's values will get better over time, and that modern societies are qualitatively better than those from thousands of years ago, rather than just different.
Hell, even here a lot of people believe that technological progress is a linear path towards the future, even when there's evidence contrary to that assumption. Without even reaching for evidence from other fields, it's trivial to find examples of entire theories that have been forgotten and rediscovered. Think of Low-Density Parity-check codes, that were developed in the 1960s and essentially forgotten for 20 or so years. A lot of technology has been found and then lost.
> Hegel developed a theory that historical progress goes always forward, or towards better things.
This is a popular supposition, but is mostly a misunderstanding of Hegel, promulgated by Left Hegelians like Marx. Hegel’s argument was that thought moves toward a greater state of contradiction, not progress, but that this evolved sustainment of the contradiction represents a more rational form.
Todd McGowan at the University of Vermont wrote a whole book on this subject:
On the one hand, limitless growth is an unsustainable anti-pattern (when it manifests in the biological realm, we call it cancer).
On the other hand, it's hard to see how civilization could continue to function without the prospect of "growing the pie". If/when we reach a steady state of finite resources, it seems highly likely that the best strategy (individually and tribally) is acquire resources at the expense of one's neighbors (before they acquire your resources first). This has been the default state of nature for nearly all biological history, with occasional exceptions of growth and plenty.
I don't have an answer here, long-term; and where the rubber meets the road, I do think our corporate model of "your business is failing if it's not growing" results in more negative than positive externalities (at least, beyond a certain equilibrium), and should be eyed critically. But the growth model is as much about social mindset as it is real-world wealth generation; and luckily we have lots more progress we can make (both on and off Terra Firma) before we're at risk of fully diminished returns on the growth of sentient well-being. The question is one of intelligent growth (cue Bucky Fuller [0]), rather than a locust-like runaway replicator pattern, which is clearly net-negative even if it looks like growth when zoomed in.
Or even if it's actually happening - it's a problem that the Hegelian concept of progression has rooted itself so deeply in the west that we think it's universally held, where other countries find a cyclical nature to humanity and history
If we want the species to get off this mudball, we need newer and better technology, also we need it to head off the climate and pollution crises and raise the quality of health and living for everyone on the planet.
As we expand to space. On Earth there is an effective upper limit to the amount of heat we can release into the atmosphere before we'd cause catastrophic warming even without greenhouse effect.
That upper limit is probably orders of magnitude higher than we’re currently dissipating. We could blanket every city and town with permanently heated streets, and - as long as we stopped pumping water vapour into the dry upper troposphere - we wouldn’t even notice an increase in temperature.
The big problems are 1) gigatons of soot coming over the north poles, permanently lowering the albedo of the planet and 2) pumping water into the dry upper atmosphere via jet exhaust, creating permanent cloud layers where they didn’t exist.
Solve those two issues, and we’d have a much more interesting debate...
Sea ice radiative forcing and overall albedo effects are on the order of 10% of well mixed greenhouse gas radiative forcing.
Short term cloud layer albedo effects are even smaller.
If we switched all our greenhouse gas emissions to nuclear, I agree there’s huge room for localized waste heat. But reducing albedo effects isn’t going to help if we keep emitting greenhouse gases.
You missed out the ecosystem collapsing because marine life has died off, and the oceans and waters becoming acidic (which will affect trade, drinkable water, rain, etc.)
> It's a fallacy to believe we will require less energy in the long run
It's hard to say without knowing your definition of long run so would you care to explain why you believe that less energy is required in your long run?
I see it as exactly the opposite, in the short/medium term we will use more energy but it will eventually go into reverse.
Maybe Jevons Paradox will intervene and you are correct, I don't know
For the record, my long run is 150+ years, if only to make sure that we are all dead.
Energy production really isn't the issue. Energy storage is.
We produce way more energy than we use. We currently do not have an easy way to store it. So why generate more than we need? Well, we need to have power available to prevent brown outs, especially right after work when everyone gets home and flips their light switch (so to speak). Second; it is actually more efficient to keep the generators fully spun up.
> We produce way more energy than we use. [...] So why generate more than we need? Well, we need to have power available to prevent brown outs
That's not how it works. If you generate more than you need, you will have the opposite of a brown out (overvoltage and/or overfrequency). Generation and consumption of power have to always be precisely matched.
> especially right after work when everyone gets home and flips their light switch (so to speak).
When everyone gets home and flips their light switch, the extra demand slightly reduces the system voltage and frequency; feedback loops on the power plants then increase their generation to match the extra demand.
> Second; it is actually more efficient to keep the generators fully spun up.
Keeping the generators running at their maximum means you have no margin for extra demand or sudden loss of generation (for instance, another power plant or a transmission line having an issue and shutting down). It's better to keep the generators running below the maximum to have a reserve for when these things happen.
It's also a fallacy to think that humans will be able to escape the gravity well in large numbers. And once we do, the ones in space will quickly become a new species as they adapt to a different environment -- assuming they survive at all.
Science fiction is fun and all, but even an ecologically ruined Earth is going to be a much more welcoming environment to live in than anything else the solar system has to offer. Meaningful exploration of the solar system will be restricted to unmanned probes and complex missions that require 100 Earthbound support staff for each human we send into space.
Continuing to live on Earth will require a big reduction in the amount of energy we use per capita. We're already above what the Earth can realistically handle, and the global population is still growing.
Can you explain in which way solar is not a "near-infinite source of energy"? I mean... there are without doubt technical challenges and all that (but arguably less technical challenges than to do the same with nuclear), but there really is no relevant limit to the deployment of solar cells if you include things like large-scale installations in the desert.
Solar depends on weather, location and battery storage. Solar cells also require massive amounts of physical area to produce the same amount of energy as a nuclear power plant. And you can't compare a "2 gigawatt" solar installation to an equivalent nuclear installation because nuclear will produce power 24/7 at peak capacity, rain or shine. A solar plant sees a huge amount of variability and only hits peak efficiency for a brief part of the day on clear days.
That technology is far more expensive than photovoltaic panels.... for example, the solar thermal Ivanpah project is now obsolete for that exact reason.
Additionally, The storage density of molten salt is far lower than the generation capacity of a reactor.
Solar panels may be cheap to produce, but they are terrible for the environment because of the complex manufacturing process. Parabolic reflectors can be melted down and reformed without huge additional environmental costs. If you're interested in the environment, this is the real benefit.
Cutting-edge solar panels have about 23% theoretical efficiency. Stirling dish systems have about a 30% theoretical efficiency and about a 25% real-world efficiency.
Solar plus battery near the equator can be sized for 24/7 power with relative ease.
But no one would do that, because we don't need power at the same rate all the time. Near the equator, where most people on the planet live, our need for energy is correlated with when the sun is shining and air con is running and people are awake.
This is why solar is better even at the mythical "baseload" than nuclear is.
Also, the base load most of us are talking about isn't created by people for the most part... it's created by industry, without which life on the equator or anywhere else would be rather primitive.
In any case, the equator is going to become unlivable in the next few decades to the point where populations in equatorial areas will drop drastically.
Do you have any data on power consumption near the equator, or where does this assertion come from?
Anecdotally, it doesn't ring true to me from my time in Thailand and Malaysia, where inefficient air con would be left on in poorly insulated houses overnight to help people sleep. I couldn't find any graphs like you can trivially find for power grids in the west.
But I did find this article[1] which indicates the record power consumption occurred at 9:35pm, beating the previous record which occurred at 10:28pm a few years prior, and both of these are times where it's all going to be coming from batteries.
But this isn't necessarily representative. Even if your reference is a book I'd be happy to purchase it, I find solar penetration in developing countries particularly interesting.
Especially if there are advancements in power distribution e.g. high temperature superconductors.
This would be a game changer for the major continents like Europe, US, Africa where you could have power coming from hydro, thermal, solar, wind etc and being shipped to where it is needed.
Those losses are on a per 100 mile basis. 2-5% per 100 miles of line (not as the crow flies mind you). Superconductors would make a huge difference if they could build a room temperature superconductor.
Exactly, when people mention losses saved by superconductors they don't mean to improve the current grid. It's to route power from the always sunny side of the planet.
No. Sorry the Mediterranean Sea separates Europe from Sahara (as well as several countries) and you can't do this practically by any means. Europe doesn't own the Sahara.
The main problem with installing solar in the Sahara and transporting it to Europe is the political instability of the region. Laying cables across a sea or ocean is already routinely done.
There's communication cables laid across the sea. There's no examples I know of High Voltage transmission with enough power to operate Europe. It's not only the instability of Sahara, but laying a cable across the sea for the transmission of enough energy to power Europe would require defense of the cables. A submarine could otherwise attack the grid and take Europe offline. This is unpractical on a lot of levels and hand-waving this as a reasonable solution is silly.
You can't run Europe on a single 1000km line running out of the Sahara over the Mediterranean. Period. Not only would a sea cable need to follow the surface of the bottom of the ocean (therefore the line would be much longer), it would experience significant power loss as it gets stepped down from higher voltages as it distributes across Europe. It's not technically feasible at all. You would lose anywhere from 20-50% from transmission. No joke. Resistance happens. A superconductor would fix this problem. You wouldn't have to send power over dangerously high voltages to save power.
You would probably start with one cable and increase it over time.
HVDC cabels are already routinely build undersea. There is a 700km long cable under construction between the UK and Noreway. The longest HVDC connection is over 2000km long in Brazil.
Superconductors have a maximum current, if you go above that current, the superconductor will gain resistance and the heat will destroy it. So you would still use high voltage transmission.
Semiconductors don't have infinite capacity. You assume that the superconductor is also practically free. Even them, building a net is still very expensive
The 5% is quoted further in the article for overall grid waste. The per mile basis is purely there to compare why higher voltages for the Samer power lead to less waste.
Not every country has access to a desert to deploy large scale solar. Most of the EU, Russia for instance, and those places do not want to be dependent on other countries.
It's probably also worth noting that photovoltaics have to be exposed to the sun to work... which means that they can't be protected by armor or bunkers, which means that they're terrifically vulnerable to terrorist or other attack.
The terrorist attack that is going to take out 1000s of square kilometres of solar panels? Surely they'd just attack a city if they had that amount of firepower? Or in the more likely case that they don't it's easier to attack power transmission than generation.
Assuming there are that many panels, it is in fact quite easy for any industrialized country to design a weapon that can destroy a square mile of panels or more with a single warhead. They're thin sheets of silicon covered in glass.
A square mile is 2.6 square km. You are replying to someone who said it would be hard to destroy thousands of square km of panels. It would seem that your reply, that "it is…easy…to…destroy a square mile of panels or more with a single warhead", is not to the point.
However, by coincidence, your position is right, for two reasons:
1. Single nuclear warheads routinely have blast radii of tens of km rather than, as you suggest, hundreds of meters. So in fact a single warhead can indeed destroy thousands of square km.
2. 1000km² of solar panels would be 1000 GWp; at a low-cost module price of €0.19/W (the average for 2019) that's €190 billion. Currently solar plants are not built that large, nor nearly so.
I was replying to a comment that said solar panels are vulnerable to terrorists.
If terrorists have a nuclear warhead they are going to use it against a city not solar panels. Similarly if terrorists have a weapon that can take out thousands of square km of panels it can easily take out existing powerplants (which btw are not armored or in bunkers - apart from some limited shield on nuclear powerplant cores.
The point here is that solar panels are not more vulnerable to terrorists than the existing infrastructure and are probably less vulnerable due to size and how distributed they are.
Most militaries rely on oil and dirty fuel to run campaigns. There's mandatory fuel stockpiling. The event of a power station going down would be seen as an act of war and suddenly there's much bigger problems for everyone involved to worry about.
What I have heard, you do loose energy in transmitting power over large distances. There exist low loss transmission cables, but I don't know how effective that is over very great distance. There is also the question of how to redesign a distributed power grid into a centralized version where all the power comes from a single sources in desert areas.
On top of that there are political challenges. I have a hard time imagine Europe in current climate being happy to rely exclusively on power from the Sahara.
The Sahara thing is ultimately just to illustrate that there's no real limit to solar energy (because the upper commenter claimed that nuclear is the only energy that is practically limitless). But of course in practice you'd do the easy things first - that is, build solar on every rooftop. No country is anywhere close to that.
I guess in the future we'll use imported solar for hard to solve problems, e.g. turn it into hydrogen or synthetic fuels (which also makes the transmission loss problem much smaller), while our electricity needs will be served mostly by local wind and solar.
There's no solar energy in the Sahara at night. And then as people mentioned, there is energy lost in transmission through resistance in power lines. There's also the fact that not too many of the world's 8 billion people live within serviceable range of the Sahara, assuming you can get past the geopolitical instability in that region to construct and maintain such things. The solar cells would need constant cleaning from dust storms to keep them running at high efficiency.
No, running solar on rooftops isn't the most practical use either. Depending on latitude, weather, cost of solar installation and battery installation, orientation and layout of roof to the sun, the problems with snow, rain, and hail, the lack of solar at night, the fact that none of this generates enough power for those times when you need it most like in the middle of winter in northern climates, etc. Solar and wind will never meet the growing needs of modern economy. Period. It's a pipe dream.
They are great supplemental sources of electricity. They cannot power a first world economy.
Every thread in HN is filled with pessimistic people pointing out how things can't work. It's tiring to read.
Solar and wind absolutely can produce all the energy the world currently needs, using only a tiny fraction of available land area. You could power the whole of the US by 100 square miles of solar panels in the southwest, backed with one square mile of batteries [1]. Clearly it's a hard problem and there are many obstacles to overcome, but just as clearly it's not fundamentally unsolvable.
Long-distance electrical transmission is actually pretty efficient nowadays, so that's not a showstopper either.
Bottom line, optimists are responsible for progress and while many people on HN are content to write comments about how it can't be done, somewhere there's an entrepreneur working hard to make it happen - and the smart money is on them, collectively, over the long-term - and thank goodness for that!
You can absolutely power the whole of the US by 100 square miles of solar panels in the southwest. The cost would be to build the solar panels, the square mile of batteries, replace all the power lines in the US with cables that can handle very high voltage (twice or more than what current power transmission cables can handle) needed for long-distance, and replace all the power stations connected to those so that can take that high voltage.
It not impossible at all, we have the technology, it just money. Replacing 200,000 miles of cables, with a price tag of a few millions per mile is a project the US could undertake. Replacing all the power station to handle the very high voltage is similarly possible.
When choosing between the many alternatives it is something which should be calculated next to the cost of building nuclear plants in a distrusted pattern, and the long term cost of nuclear waste that such plan would entail. If entrepreneurs could invent power transmission cables and power stations that can manage millions of volts and cost a fraction of existing methods to install would make a centralized place for power generation a much more attractive option.
It's a lot more than 100 square miles. Here's an NREL report on this exact question: https://www.nrel.gov/docs/fy13osti/56290.pdf If you extrapolate (see analysis here: https://www.freeingenergy.com/how-much-solar-would-it-take-t...), you get 21,250 square miles. Elon musk says 100 miles x 100 miles, which is 10,000 square miles. Both of those figures are just for current electricity demand, so if we electrified all transportation and industry, we're looking at maybe 60,000 square miles.
60,000 square miles is half of Arizona. Now suppose we want to scale up energy consumption in the U.S. by a factor of 100. At that point, you're at twice the land area of the U.S. Even at current consumption, the amount of ecological damage you're causing by covering half of Arizona with solar panels is huge.
On the other hand, if you wanted to replace all U.S. energy production with nuclear, you'd need about 7-10x more than we have today, or about 1000 reactors. The land area for these reactors is about 700 square miles, or about 25x25 miles. If we wanted to scale it up by a factor of 100, we'd be looking at half of Arizona again.
Sorry, it's 10,000 square miles in the article I linked. I can't edit my post anymore, but my point stands.
Energy consumption is actually dropping in the US currently, but I don't expect that trend to continue indefinitely, eventually we will get to 100x energy usage. And you're right, at that point solar wouldn't cut it.
But that's a far cry from your original post saying it won't work for a developed nation, when it clearly can work for the world's richest and most energy intensive nation.
As for the very far future when solar won't cut it? I'm sure we'll use a lot of nuclear, and by then probably a lot of nuclear fusion. Or maybe we won't have those crazy energy requirements because we've moved most industry off the planet like per Jeff Bezos' vision of the future. It's enough to get ourselves sorted in the present, so we can have a bright future, and solar and wind power can help us achieve that. There's no reason nuclear can't be a part of that picture, but they have a hard challenge ahead because current nuclear is not competitive cost wise with renewables plus energy storage.
The solar panels do have to be fabricated. That probably doesn't have the same scaling potential as nuclear given how much power/m2 nuclear can reach.
Also, at a guess the energy in solar panels drop with the square of distance to the sun. It is unlikely to be a good choice for interstellar travel if 'advance[ing] as a species' heads in the more fantastic directions.
If we are getting to the point where W/m2 is an important win for nuclear, we are going to start to have serious problems with the rejected heat.
A fundamental limitation of thermal steam engines is that they can only ever be 50% efficient, and you have to dump that waste heat in order to maintain power.
Already, heat mitigation systems for some existing nuclear plants are starting to fail during heat waves as the climate warms. And these are expensive systems: at Diablo Canyon in California, it's cheaper to replace an entire, functioning reactor with renewables than it is to simply build a new cooling system.
Which is all to say that nuclear won't scale tremendously well unless we 1) figure out fusion, and 2) figure out direct conversion of energy to electricity rather than using steam turbines to mechanically drive a generator.
For the ultimate goal that many people have for nuclear, as a power source when not on earth, these sorts of advancements are also likely also necessary. Cooling in space is not a trivial matter.
Solar power is simply indirect nuclear power anyhow. If we can do what’s happening in the sun in a bottle - then no need to produce panels. If there is material in stars to produce solar power, then there is more potential nuclear power. I’m for both, and solar seems distributed in a way that makes me think it’s better for human society, but since stars are made of fissable materials - by definition there is more “potential energy” in fission than solar, since solar is simply a subset of fission at a distance.
Solar is fusion and not fission [1]. If you can't even get that right, why should anyone take these weird pro-fission arguments seriously in 2020?
The solution is to keep using existing nuclear power and develop renewables for replacement. Nuclear fission plants take at the very least 10 years (!!) to go online from the day construction begins. And that leaves out years of planning and dealing with contracts.
It's too expensive, dangerous and redundant in the face of emerging renewable tech which is becoming cheaper and more efficient by the month.
Renewables are not a replacement for existing nuclear power unless you either add fossil fuels or batteries to the mix. Countries which currently are replacing nuclear power do so with a combination of renewables and fossil fuels, with fossil fuels burning when renewables are not producing.
Batteries, usually reverse hydro power, is an interesting future technology. Some argue it is significant more developed than fusion. The bigger question is if its economically competitive compared to fission. There is costs and energy loss in every single step of producing electricity from renewables, transmitting it to the battery, converting it into potential, recreate the electricity, and finnally transmitting it to the end users. With fission you go directly from the power plant to the end user. Reverse hydro power plants also take a long time to build and either use a lot of land or coast. If you build it on land it also release a lot of methane as top layer of the land decompose.
Countries which currently are replacing nuclear power do so with a combination of renewables and fossil fuels
Which countries? Germany for example isn't - yet. We're still in a place where we can reduce usage of both fossil fuels and nuclear, though that won't last unless we figure out effective means of energy storage.
As an example, Sweden. People will use fossil fueled energy when the choice is between people freezing in their home or burning fossil fuels. Sweden rely on a mix between hydro and nuclear, but it is not feasible to extend hydro beyond current capacity. The nuclear plants however is getting older, and politically people want to shut them down. Something has to produce the energy, and during the winter it is imported fossil fuels energy when the wind is not blowing.
Germany as an example illustrate the issue quite nice, as can be seen live at electricitymap.org. When the wind is blowing the country goes green with around 70% of energy being produced by wind. Very sunny days you get around 20% solar. Days like today that is a bit rainy and not very windy, and you have 60% fossil fuels. The constant is nuclear around 10%, so remove that and the above numbers will go up depending on weather conditions.
German anti-nuclear activists like to tout the percentage figure of renewables in the country, but that's not the right metric to look at : the coal+gas baseline is so bad in terms of CO2 emissions, that even in ideal conditions when wind production is 70%, German electricity's carbon intensity is still way higher than in France, Sweden or Iceland (
Coal causes 35.000 premature deaths in Europe every year, and 7 of the 10 most polluting industry sites on the continent are German lignite power plants.
The hypocrisy and constant lecturing from Die Grüne needs to stop.
Sure. But that doesn't make Germany an example of replacing nuclear with fossil fuels. From 2002 to 2019, percentage wise, fossil fuels went down by 1/3 (from ~60% to ~40%) and nuclear by 1/2 (from ~30% to ~15%).
Since electricitymap gives current say 40% coal and 15% gas, to a total of 55%, I assume fossil fuels are not down to 40% all the time. What you are describing is the average.
Feel free to prove me wrong, but when the wind over Germany is still (<4ms) and its night, the amount of energy production using fossil fuels are higher than 70%, and thus at peak, fossil fueled energy production is higher now then before when nuclear stood for 30%.
and thus at peak, fossil fueled energy production is higher now then before when nuclear stood for 30%
But that's an irrelevant metric: What matters is the total CO2 released, ie the integrated value. So short-term, you replace coal plants by gas peakers, and transition to next-gen storage mechanisms long-term (better batteries, cryogenic storage, power-to-gas - the latter is especially interesting as Germany has pre-existing gas infrastructure than can store hundreds of TWh, and we use natural gas anyway for heating and industrial purposes).
What matters is to close down the fossil fuel plants for good. The output of wind is on average 45% of maximum capacity for offshore wind, meaning if you have 100% of wind during optimal conditions and 0% when the wind is still on average you get around 45%. For land based wind parks the numbers are low, around 25%.
So far there is very little investment to build out wind beyond having 100% wind in optimal conditions. Germany has almost hit that point, and if we look at neighbor Denmark then we can see what happens when it does reach 100%. Building wind beyond full capacity turns uneconomical quickly, as investors found out in Denmark.
The result is that the coal and natural gas plants will burn and continue polluting the world. The competitiveness of renewable is based on the cheap initial costs while it goes towards max capacity. The price tag does not include overcapacity, the batteries, cryogenic storage, power-to-gas and so on. It works fine as long as we don't think about the fossil fuels that get burned when the wind is still.
Nuclear plants have a linear cost. Going from 10% to 20% cost just as much as going from 90% to 100%. No overcapacity, no batteries, no conversion loss. You add 10% nuclear plants and you can demolish 10% fossil fueled plants. You build 10% additional wind farms and the same old fossil plants must remain. You demolish 10% of the nuclear plants, and you have to build the same amount of new fossil fueled plants in order to compensate when the wind is not blowing. New fossil fueled plants are going to get used, investments is going to be repaid, and political influence fill make sure that they continue to operate.
> What matters is the total CO2 released,
If people really thought so they would look at the electricitymap and look which countries does exactly that. Who has the lowest total CO2? The answer: those that can produce a constant base load without releasing CO2. Hydro or nuclear. Those that have invested most in renewable are not the ones with lowest total CO2.
> "Nuclear plants have a linear cost. Going from 10% to 20% cost just as much as going from 90% to 100%. No overcapacity, no batteries, no conversion loss."
In your nuclear-only scenario, without storage you'd need enough capacity to cover peak demand. This can be 2X or even 3X higher than average demand, so there would indeed be significant overcapacity. Very expensive!
Typical nuclear plants are also not good at demand response: to operate efficiently, their output must remain constant most of the time. Over-capacity at off-peak times is potentially a big problem on grids with a large portion of nuclear.
Some combination of storage, peaker plants, and demand response is required regardless of whether nuclear or renewables are used. The most cost-effective future grids are likely to use a diverse mix of technologies.
Germany absolutely is relying on coal and (Russian) gas to afford its ideological decision to prematurely sunset nuclear plants, at the worst possible time in history : just as climate change becomes an emergency.
The cold hard truth is that it's impossible to operate a grid with solar & wind energy alone, unless and until a hypothetical battery storage breakthrough lands in the next decades.
I've just checked the realtime figures and as I write this, German electricity is 5 times more carbon intensive than in France (72% nuclear) : https://www.electricitymap.org/
New-build nuclear is far more expensive, per kWh, than renewables. In Europe, even off-shore wind - one of the most expensive renewables - is now coming in much cheaper than nuclear projects.
In the US, even many old nuclear plants are struggling to compete without subsidies against renewables and natural gas.
> "Nuclear fission plants take at the very least 10 years (!!) to go online from the day construction begins."
Yes, but that's what the small modular reactors being proposed by Rolls-Royce, and others, intend to solve. If successful, they would greatly reduce the construction time, risk, and cost of nuclear projects.
It's a great project for the technology alone but isn't the projected time frame too late? Where will we be 10 years from now with renewables?
Also, if Rolls-Royce projects 2029 it doesn't mean it's done by 2029 and most certainly not wide scale deployed/operable. So what kind of renewable infrastructure and tech will be deployed 15-20 years from now?
> "It's a great project for the technology alone but isn't the projected time frame too late? Where will we be 10 years from now with renewables?"
It's not a question of nuclear or renewables - we absolutely need renewables, and right now renewables are much cheaper, and can be delivered faster, than nuclear.
But there are regions of the world that may struggle to decarbonise completely without nuclear in the mix. Especially if you consider additional demands in the future from electrification of transport, building heat, etc.
> Where will we be 10 years from now with renewables?
If the electricity storage problem doesn't get solved (which is a pretty small "if", since it requires a very uncertain breakthrough in physics) : nowhere.
In 10 years the climate emergency will be even more salient, but one of coal/gas/nuclear/hydro will still be required in the mix.
Countries that can't have hydro for geography reasons, and have shut down nuclear early for political reasons will be a liability to the rest of the world.
Most likely in that time frame we'll be trying to get by on a mix of renewables and non carbon neutral generation from coal or similar, and we'll also have implemented limits on carbon generation that will effectively be crippling various industries and increasing the cost of various necessities worldwide.
We won't have a choice... it'll be down to either everyone accepting reduced quality of life or nuclear... at which point nuclear starts to look very good.
>It's too expensive, dangerous and redundant in the face of emerging renewable tech which is becoming cheaper and more efficient by the month.
Why are you comparing the state of nuclear energy today with the potential scientific breakthroughs of renewable energy in the future?
If you compare nuclear of today with renewables of today, then the winner is clear. If you compare the two accounting for potential scientific breakthroughs..who knows?
Stars are fusion reactors, not fission. They're mostly made of hydrogen and helium, which account for 98% of its mass: http://hyperphysics.phy-astr.gsu.edu/hbase/Tables/suncomp.ht.... You're going to have a hard time building a fission reactor running on hydrogen and helium.
Stellar fusion like that which occurs in our sun is effectively aneutronic, but is also relatively slow, because the limiting step is the combination of two protons into a proton-neutron pair.
You need to keep a lot of hydrogen at plasma-hot temperatures and very high pressures for a long time. So you can't really do it with masses smaller than Jupiter, because smaller bodies can radiate the energy away faster, and produce fewer events from the lesser mass.
So the only technologically effective way to leverage solar is to deconstruct larger stars into red dwarfs between 0.08 and 0.35 solar mass, perhaps with a ferro-platosmiridium core to increase the overall density and make the reactions viable at lower overall mass. Then surround the whole thing with a Dyson shell and Shkadov/Caplan thruster.
Aneutronic reactors are good of course, but the Sun still puts out lots of other dangerous radiation. In particular, ultraviolet electromagnetic radiation from the Sun causes tens of thousands of cancer deaths every year. I have no problem accepting a few percent of that mortality from man-made radiation sources if that helps solving the CO2 emission crisis.
Good. For the foreseeable future, the limit of energy available from solar will be the total surface area of the Earth divided by two.
And that includes hydro from evaporation-rainfall cycling, photosynthesis, and wind. Which basically leaves as alternate energy sources tidal, from the sun and moon dragging the oceans around, energy stored from long periods of solar absorption in ages past, residual geothermal, and nuclear.
A little neutron-activated waste is indeed a small price to pay.
Forgive me if this is a worn trope, but I've read somewhere that the upper bound on what we can do with nuclear energy is not how much we can produce, but what we can do with the leftovers ... as more and more people use nuclear power we're going to produce more and more nuclear waste, and there isn't really any answer as to what we can do with that. There are plenty of fantastic "ideas" as to what can be done, but it doesn't seem to me that we've fully solved that side of the equation, and I'd be thinking we really should have a good bit more work done before we consider something with such dangerous side effects "a solution".
Nuclear waste is significant less harmful than fossil fuel waste which currently is produced in massive amounts.
For my perspective, we should ban the worst waste first and then iterate. If we can build an energy grid without burning fossil fuels we should do so, preferable yesterday. If we can then build one that also is without nuclear waste then lets do that too, but my first priority is going to be to get rid of the fossil fuels.
What I do not want is replacing nuclear waste with fossil fuel waste. While we have an unsolved problem with nuclear waste, it is dwarfed by what can be done once run away climate change happens. A world where 100% of energy comes from nuclear is preferable over one where 100%, 80%, 50%, maybe even as low as 20% comes from fossil fuels.
If literally all the electricity ever produced by the human waste had been produced by current generation nuclear technologies, there would be a small hill of high-level nuclear waste somewhere. It wouldn't be an issue except locally where it was stored.
To believe that is a problem is to not have grappled with just how big the world is and how much of it is uninhabitable to humans already. The human population is concentrated in an absurdly small footprint in major cities and fertile belts compared to the size of the planet. The area the waste would sterilise would be a non-issue.
I dunno. What do you want to be solved? If we call it poisonous instead of radioactive would you be happy? There are literally poisonous lakes out there and nobody cares much. One more doesn't matter. The only interesting thing about nuclear waste is we use a different word to describe the same outcomes. The outcomes don't seem that dangerous in the big picture.
> If literally all the electricity ever produced by the human waste had been produced by current generation nuclear technologies, there would be a small hill of high-level nuclear waste somewhere.
I'm not sure what kind of "modern nuclear technology" you're referring to. Are you saying that our legacy power plants are bad, and should be replaced? At what cost?
> It wouldn't be an issue except locally where it was stored.
So, a single nuclear power plant for the world?
Transporting nuclear waste is also a problem. Even in the US, where it doesn't need to cross oceans (ignoring Hawaii, Puerto Rico and maybe some other territories).
I'm also uncertain if we'll be likely to encourage modern nuclear reactors in Iran, North Korea and in various failed states. They may be safe wrt weapons grade nuclear weapons initially - but could the be modified? (honest question, I'm not sure how easy it would be to enrich material for a traditional bomb, or indeed a "dirty bomb". But small amount of high grade waste kind of sounds like it's usable for a dirty bomb?).
Most of the nuclear waste that exists is from weapons production. Nuclear generating plants produce only a very small amount.
Right now, small enough amounts of waste are produced that reactors generating power actually store the stuff on site.
>but could the be modified?
Modern reactor types are specifically designed not to be proliferation risks. The only reason the older reactor types are risks is because the governments who originally built them wanted to produce weapons, so they chose the technology that allowed them to do so.
Waste could be used for a "dirty bomb" in some sense, but it wouldn't be terribly effective. "High level" is relative, and the isotopes that would make a dirty bomb truly scary aren't available except in fuel rods shortly after their removal from a reactor... at which point no one does anything to extract those isotopes anyway, they just stick the fuel in cooling ponds to decay down to lower levels of radiation.
Yeah man, I dunno either ... I think "poisonous" is understating it somewhat. We're talking about substances that will remain toxic for thousands of years, and are quite happy to go everywhere they can if there is some issue with containment.
I appreciate that you have a conviction that this is a problem, but you're coming across a bit hand-wavy in your arguments. I'd prefer to see concrete solutions (and I don't mean nuclear waste encased in concrete) than rhetoric as a means to address my concerns.
> you're coming across a bit hand-wavy in your arguments. I'd prefer to see concrete solutions
It is an order-of-magnitude argument; a bit like arguing whether $1 billion or $1 million is more dollars. The difference between the two figures is almost exactly a billion dollars because there really is no comparison between orders of magnitude. Uranium is something like 6 orders of magnitude more energy dense than fossil fuels (so more of a trillion to a million) - the waste is a lot worse too, but it is nowhere near 6 orders of magnitude more dangerous, because that would suggest it is killing more people than the population of the earth already. Which it is not ^.
You can say you want something solved, but the problem you want solved is several orders of magnitude smaller than the problems everyone currently shrugs off as totally normal. The orders of magnitude are so different they do not need to be solved and can be handwaved. The nuclear waste problem is incomparably small compared to the fossil fuel problem which has proven to be tolerable despite 20+ years of resistance by Green groups.
It is also probably going to turn out to be smaller than the waste problem fabricating renewable will have by the same order of magnitude issue.
^ The evidence suggests it is actually not that much worse because it is so easy to isolate. It is practically achievable for nuclear waste to do less actual harm unit-to-unit than coal.
You're comparing magnitudes there ... but I've a feeling there's a base-rate somewhere, relating to how dangerous just a relatively small amount of nuclear waste can be if it gets into groundwater or something.
The real problem with nuclear is that it's a one way only system. The effects of other forms of fuel can in theory be sequestered eventually. Sequestration of nuclear waste is exactly something that yout don't want to happen.
As you say, it's a matter of scale. A limited amount of nuclear power is probably fine, and safe. But it can never be the "solution" to our energy problems until the various problems are solved satisfactorily.
We could grind nuclear waste from current nuclear plants into fine powder and intentionally blow it into the atmosphere and still cause fewer deaths than coal, as well as cause the release of less radioactive material, as the coal industry causes huge amounts of uranium dust to be released in the air.
So as it stands, if we look at the real world instead of some hypothetical future, we continue to depend on types of power that causes not just the release of more harmful material, but the release of more radioactive material than nuclear.
If we get to a point where we have fully supplanted fossil fuels, and we need to consider whether to continue building nuclear or replace it with alternatives, then the situation may look different, but at the moment anything that slows the replacement of things like coal causes massive amounts of harm, both environmentally and in killing people.
We could have a Chernobyl a year, and it'd still cause us less harm than the continued dependence on coal.
> We could grind nuclear waste from current nuclear plants into fine powder and intentionally blow it into the atmosphere and still cause fewer deaths than coal, as well as cause the release of less radioactive material, as the coal industry causes huge amounts of uranium dust to be released in the air.
How does the math work on this? It seems... hyperbolic.
Coal contains traces of uranium, which gets released during burning. The uranium released in such way would have been sufficient to generate the energy obtained from burning coal if it used for fission instead [0].
So indeed, now we disperse the nuclear fuel in the atmosphere and freaking out about it much less then when instead it is being processed in a plant, and coal left alone.
Look up fly ash. Orders of magnitude more fly ash is produced than waste from nuclear plants, and it contains high concentrations of uranium and thorium.
While plants in e.g the US now captures most of it, huge quantities are still released into the air especially in countries with lower environmental standards.
But even places where it doesn't get released that just means having to deal with far more radioactive waste than nuclear plants produces.
Naturally occurring radioactivity in coal ash means that coal burning power plants release far more radioactivity into our environment than nuclear plants do. In fact, if you want to get the least amount of radioactivity into your body, the safest place is behind the shielding of a nuke plant, because it would also protect you from naturally occuring radiation.
If you add up the total of radioactive elements in Bequerels released annually by coal plants and then assume 100% of all waste from power generating plants could be ground up and released and count that up, the amount of radioactivity from nuclear plants would still be less than coal.
We burn a LOT of coal, and despite the media's portrayal of how much a problem radioactive waste is, it's very overblown for power generating plants. Weapons production is another matter, but we've already been trying to stop that from happening for years.
There exists coal with more fissile energy, in the form of heavy isotopes such as thorium and uranium, than it has chemical energy, in the form of carbon-carbon bonds.
This is then burned in power plants, releasing the radioactive material into the atmosphere, and leaching it into groundwater from exposed piles of fly ash.
Fast reactors as well as other specialized designs can burn down fuel that is currently stored as "waste", and just don't produce such long-lived isotopes.
For examples, the half-life of output products from uranium-fueled SVBR-100 is ~550 years, and that can be reduced further by several technologies that are now available.
>We're talking about substances that will remain toxic for thousands of years
Most substances are toxic forever. If you bury mercury or lead in a hole a dig it up in a few million years it will be just as toxic. Radioactive substances are an anomaly in that they become less toxic over time.
Nuclear energy has a PR problem that is very hard to solve. It doesn’t help that movies dramatize it. From what I remember, the only people who died due to Fukushima were those that died of fear. The reality is that nuclear is remarkably safe, clean, etc.
By human emotional calculus, it is better for a million people to experience on average 0.001 of a mortality event than for 1000 people to be killed outright.
When we measure such things in terms of "increased cancer risk" and "possible thyroid dysfunction", it is far easier to discount, particularly as young people are not tremendously concerned when people above a certain age die of diseases that are already typical in the aged.
So even if we knew it exactly, people would still care less. It might be more relevant if, instead of death toll, it could be given a money value derived from additional healthcare expenses for exposed individuals, because young people implicitly know that they are the ones who pay when old people get sick.
Same thing for the death toll from coal burning plants. All that ash, deaths from that radioactivity and lung disease, deaths caused by global warming, etc.
If you look at a long enough time scale, anything can seem like a giant problem.
> From what I remember, the only people who died due to Fukushima were those that died of fear. The reality is that nuclear is remarkably safe, clean, etc.
Then you have a faulty memory, and a selective one at that because the crisis is still on-going; there were an estimated 2000 from evacuation alone:
You think this is safe or healthy? 100k+ displaced people living in abject squalor in the 3rd richest nation on Earth? Often seen as less-thans by their fellow citizens due to the Meltdown:
What's even more conflicting is that this year's Olympics are scheduled to take place in Tokyo, all the while the food is contaminated, as is the water (and the air if they're still doing regular debris burns that spreads it around the World).
The cancer rates, thyroid maladies and heart disease are all correlated to the radiation exposure, but they don't have an interest in monitoring this accurately and reporting it to the Public due to typical Japanese 'cultural norms.' So, in it a very defying sense of abnormal behaviour, Japanese house wives have taken to measure their neighborhoods, as well as the food and the vacuumed debris.
This is quite honestly a bigger part of why Humanity has to solve its energy crisis, Greta makes a good case for what their generation is left to live with, but being in between the two generations as a millennial and having been around for both Chernobyl and Fukushima, its hardly comprehensive of the true costs. That last video even delves into the Children of Chernobyl, they are reporting large frequencies of cancer and various immunological diseases. This is more the norm that I ever thought in surrounding areas, when I lived in Croatia it was also the same. When I lived in Germany their were patches of Earth that looked scorched that had been hit particularly hard due to the Fallout of Chernobyl. Many farming families in that area went Bankrupt due to it.
I honestly think people like you should only be able to have this opinion if you live near Nuclear Plants, for a decade at a minimum. You'll see first hand how perilous it could be, the infrastructure around coastal areas is another bottle neck that most don't consider an issue for things like evacuation until its too late; they often only have 1 way in-1 way out layouts.
Nuclear regulation is a joke, and is as entrenched and as corrupt as Big Oil. The legal system, in both Japan and the US, is equally as complicit as the Nuclear lobby and refuse to take preventive action, as was the case with why Fukushima was left exposed on the coastal area after TEPCO was warned, repeatedly by several studies, that is was prone Meltdown should something like that Tsunami happen. The Nuclear village/TEPCO/Japanese Government did nothing:
There were no known deaths from accute radiation syndrome[0]. There were deaths from things like the tsunami, the evacuation, and stress, which was essentially the point I was trying to make. The nuclear part of the equation caused a lot of fear, but it was all of the other things (and maybe the fear itself) that caused the deaths. That said, you're not wrong about the long-tail effects being hard to quantify and measure.
> I honestly think people like you should only be able to have this opinion if you live near Nuclear Plants, for a decade at a minimum.
For what it's worth, I've lived near nuclear power plants for over 30 years, and have no problem living near them until I die-- which will almost certainly not be from radiation unless we have a nuclear war. I had more radiation exposure from the coal fired power plant in my childhood town than from any of the nuclear plants I've been around.
> we're going to produce more and more nuclear waste, and there isn't really any answer as to what we can do with that.
There are several answers to the question "what we can do with it", such as 1) reprocess it and use it again; 2) keep it in the power plant pools or similar storage facilities; 3) dump it to some deserted place where it isn't a big problem (high depth, stable earth crust). In the past, UK just dumped nuclear waste in barrels into the sea, which seems kind of convenient and irresponsible, but if done right (better isolation from sea creatures), this could work too.
It is true that there is no single universally agreed upon answer. But that is the same as with all other waste. Most of waste gets either burned or dumped at some place. The same will happen to nuclear "waste", until people start reprocessing it.
I didn't think nuclear waste could be "burned" ... I was given to believing that containment was the only option right now.
I'm familiar with your (1) (2) (3) items, but again, from what I've read these aren't fully satisfactory. (1) is probably ideal but hasn't really been cracked, (2) and (3) are just different facets of containment, but (3) is admittedly the most plausible right now.
We can tolerate a limited amount of this for sure, while we work on other solutions, but unless this question gets resolved it will hamper the widepsread adoption of nuclear.
The UK approach is interesting because yes, they just dumped it in the Irish sea. There's a deep underwater ravine between Scotland and Northern Ireland where it's all dumped, along with various other bits of old military hardware and other bits that are inconvenient.
Think about that the next time you here Bojo talking about building a bridge to Northern Ireland.
Fast reactors can burn waste pretty well, as few other designs. However, it's politically problematic because it involves using plutonium burning (not as fuel, but produced in the reactor while burning down uranium).
This is not liked by certain governments, even if theoretically NPT gives a framework to do it safely, and large scale commercial reprocessing essentially died after India used Canada-built CANDU reactors to kickstart their nuclear weapons program.
Global-politically, there shouldn't be any problem for the US, UK, France, Russia, and China to have fast-breeder reactors, but it still may be a concern in local politics due to environmental or terrorism concerns.
Nuclear waste has to be shipped to the facility. All the shipping routes from all the nearest waste-producing reactors converge there. That's naturally a concern to all those who live nearby.
A quick google search says if we only used nuclear we'd get 40g of waste per person per year assuming western energy consumption standards[1].
Multiply by 8*10^9 people (a little more than the current world population) and you get 320,000 metric tons which is about 3/5th the capacity of the largest oil tankers.
Finding a place for that much waste per year is a political problem, not a technical problem. There's plenty of geologically "safe enough to outlast the radioactivity" places we could dig a deep hole (thanks to the fossil fuel industry that is a solved problem) to dump that much waste into.
Considering the already realized catastrophic disasters due to failures of our fossil fuel transport system (Exxon Valdez comes to mind) it would still be a net improvement.
Not really. The supply of economically extractable fissile material is limited in just the same way as any other mineral.
If energy prices rise then it will become worthwhile to extract, if they fall then it will be less so. Of course if the number of consumers rises this will also improve the viability of mining. According to this Wikipedia page: https://en.wikipedia.org/wiki/Uranium_mining_in_Australia, it is currently uneconomic to proceed with several mining projects.
Regardless of economic considerations fissile material is a finite resource although it could be that we will never reach the limit.
In practical terms, however, switching nuclear reactors to use the Thorium fuel cycle would allow us to use a supply of fuel that would probably outlast human civilization.
Id he happy if we just focused on social issues and happiness. But might is right and every nation state and company and human being needs all the technological might we can muster in order to “win” (win what!?)
Scientists are not always thinking through consequences in the real world though. Just the theoretical aspects of something.
Theoretically I find nuclear fission impressive and amazing.
Practically I see no desirable outcome without bad effects especially when coordinated by the private sector with their habit of saving money and squeezing out more profits every year.
> private sector with their habit of saving money and squeezing out more profits every year
Bingo. I have no issues with nuclear power. I have an issue with the human component. Safety regulations have been continuously changed and been made more relaxed to keep up with the crumbling nuclear plants that are way past their design life and coming up on another extension.
Have a crack in the concrete? Increase the allowable crack tolerance. Have leaking radioactive water? Change the way the test is performed so it allows you to still pass.
No that is an absolutely terrible way of thinking about ANYTHING. That is what the townspeople said to the aristocrat funding development of the microscope. They demanded he care for the sick before engaging in this glass grinding business which had no obvious benefit
Yet the microscope proved to be the biggest advance in medicine. Nobody could have predicted that being able to look at really small things would have an impact on how we understand disease.
Likewise nobody could have predict the profound impact satellites have had on our economy and on science.
Space exploration brings technological progress. If you are concerned about waste of money then first start to worry about industries we spend far more money on 1) Gambling 2) Cosmetics 3) Weapons, bombs etc 4) Disposable fashion
Not to mention most of the worlds problems are of political nature. They are not problems a scientists can solve. Insisting that a physicist should apply his skills to create say world peace is a waste of skills and effort.
The microscope story a sibling comment mentioned is part of a 1970 letter from a then-NASA Director to a nun serving in Africa who asked the same question. I believe the full text is worth a read:
http://www.lettersofnote.com/2012/08/why-explore-space.html
I think space colonization is an even bigger fantasy than relying on "future tech" that will magically terraform the earth and reverse climate change. We can have a tiny structure housing 10 or 20 people somewhere in our solar system for the price of millions of pounds of fuel and billions of dollars.
Inter-system travel will never happen. Forums like this tend to have a healthy population of the sci-fi minded, so it won't be a popular opinion here, but the laws of physics simply rule it out. And the "men once thought they couldn't fly" argument doesn't carry water with me. We know a lot more about what we don't know now than we did then
Nevermind the hocus pocus, and I agree as far as "travel", but all that's really needed for (fairly dystopian attempts at) space colonization is surely artificial wombs?
Which aren't far off.
We don't need to break the laws of physics, or have cryonics or singularity shit work out to eventually consume the universe. All the nodes will just be real isolated.
There will always be some mess on Earth and anywhere else human kind visits in the future. We are flawed. Let's dream about perfect, but settle for good enough.
That's like saying "let's first neatly reorganize all the files on these 700 harddrives before thinking about backups" when you can just do both.
Spending for space exploration is currently at a laughably low level, if countless people didn't manage to solve all our problems in the course of decades then adding a few billions of dollars won't do anything except make a few people even richer.
You're missing the point. Going to space puts us in very restrictive environments and situations that forces us to solve issues that we "can" just shrug away here on earth (meaning we move it to one or two generations down the line so it's not our problem anymore).
I'm amazed that the industry is still propagating the mistake made in the early nuclear age of using PWR and solid fuel. The more I read about nuclear the more convinced I become that SMRs for next gen plants are safest and most efficient using the Thorium-U233 fuel cycle and employing molten salt to transfer heat. It's automatically safe since if a runaway reaction takes place the fuel will drain into a tank and reactions will stop. This happens if no action is taken so all the humans at the plant and all the machinery could be incapacitated and a meltdown automatically stops. It uses a cheaper fuel of which we have a virtually endless supply of and enrichment is not needed as it is with U238-U235. There's less waste as well and it's less problematic waste (decaying in hundreds of years instead of millions). Operationally it's easier to load up fuel and wastes are easier to remove. Is it just "this is how we always do these things" or is there something else going on here?
> Is it just "this is how we always do these things" or is there something else going on here?
That tech is known, well designed, approved and extensively proven through many thousand years manwork experience worldwide. It’s also the most affordable in liberalized markets. Going smaller & modular is the real novelty in the industry, for grid-agnostic coupling and decoupling. FYI: I have been consulting and designing innovative or exotic NPPs for years and I am now pretty confident I will not live long enough to see any one of them built and working, maybe not even at the demonstrator stage... but this side of the story might come good for another post ehehe.
I'm just a layman with enthusiastic interest. Can I ask you about one thing in this pipeline of decision making since you are in the field?
Refueling is easier to do in MSRs and nigh takes away an embarrassing long downtime while the solid fuel is loaded. Why is that not a first concern (less downtime means more energy from roughly same capital expenditure)?
Safety is critical. It also costs a lot. With MSRs you need less land and can place the plant closer to the point of use (frozen plug to drain tank makes it so that any runaway is contained without intervention). It's even been suggested that for high-thermal uses like steel plants you could get better efficiency by using a thermal plant instead of making electricity for at least a twofold gain in power use. Is there anything inherently wrong with that idea?
Is the dialysis design as solid as it looks like on paper? Are there operational problems with it that the advocates are not talking about?
Good to have interested laymen, really, so thanks!
The fundamental point is (in my very humble opinion, mind) you may still have a business case for nuclear in 2020 but no feasible return on investment against non-nuclear sources for decades-long experiments that, all in all, are aimed at marginal improvements in the grand scheme of things and the enormous hassle proponents have to face.
It is fusion or nothing, apart of a couple of Gen IV that are advanced enough already, for the next 20-30 years or such in the West (China or Russia might be more ambitious, but they are not liberal or neoliberal democracies, so they can play a different game to some extent)... until all the PWRs still operating worldwide must be put down for obsolescence.
For a layman computer science comparison, think of the still up-and-running fortran legacy in some sectors (banks?) you won’t really really want to drop if not really forced to... because it works, you need no fix, you do not expect any unheard catastrophic incident or accident anymore.
That's the thing. Thorium doesn't sound to me like an incremental change at all. It sounds like a gamechanger. There is no blast radius. At all. Criticality in that sense is not possible. There's improvement along every measurable criteria of operation. Some of those improvements are not small either. This is all contingent on me not being mistaken in my assumptions. I've only ever seen the argument from the advocates.
It sounds a little apocalyptic putting all of ones hopes on a technology that may or may not work at all. We have no guarantee that fusion will provide a return. It's the tech of choice because of wishful thinking. Gen 4 and liquid fuels are proven to work and eliminate a big slew of the disadvantages that made people fearful of NPPs. IMO if a part of the solution to CO2 is DAC then we will need massive energy generation to run the capture plants and we need them soon. No tech can deliver that in the next 10 years safely except Thorium.
There's nothing wrong with old tech in computer science. You can containerize Fortran batch programs with ease. All old things come back once in a while. Functional programming is now the rave and Common LiSP actually sees use.
> There's nothing wrong with old tech in computer science. You can containerize Fortran batch programs with ease. All old things come back once in a while. Functional programming is now the rave and Common LiSP actually sees use.
You got it. May I second that as a nuclear eng for nuclear? Small & Modular is your container... and the novel, distributed grid at transnational level is your framework!?
I don't think that translates well here. There aren't very many new projects being started up using Fortran. There are some. But the vast majority of Fortran is shrinking legacy because we have better solutions to those problems today (R/Python for Fortran).
Agreed that is a significant gating factor, but the situation is not impossible. Several paths are laid out for moving forward in a 2018 report by a couple of ANL scientists. The report is quite readable, even if you know little about materials science.
There are a bunch of places I've found where the difficulty (and therefore cost) of working with titanium is really holding humanity back. If we could find cheaper techniques for working with titanium it would revolutionize a lot of industries, as it's plentiful, light, strong, non-reactive, and melts at high temperatures. But the same attributes that make it hold up under the stresses of use, make it very hard to shape.
A radioactive fluoride salt might just be one of the most corrosive substances on the planet. Titanium might hold up a little longer but not by much.
Any material that can resist the corrosion will be bombarded by neutrons until it is transmuted into another material that can't, even with "slow" neutrons.
The Wikipedia article[1] says, "A 2011 MIT study concluded that although there is little in the way of barriers to a thorium fuel cycle, with current or near term light-water reactor designs there is also little incentive for any significant market penetration to occur. As such they conclude there is little chance of thorium cycles replacing conventional uranium cycles in the current nuclear power market, despite the potential benefits." What did they know at MIT in 2011 that we don't? Or were they just wrong?
If really no material will work for this, I can only imagine some sort of mag-lev design that keeps the molten salt out of contact with other materials, but that seems kind of far-fetched.
The solution is to replace the reactor chamber far more often than you would in other reactor designs. It's not actually an insurmountable problem because fluoride chemistry is extremely common in nuclear engineering - most of our nuclear waste is stored as uranium hexafluoride after all - and there is a lot of expertise in dealing with exotic corrosive materials. It's just that replacing your entire reactor every year is astronomically expensive. Just because the technology exists to do it, doesn't mean that it can be done economically when competing against natural gas, solar, and wind.
And we understand it to be rather dangerous in failure modes. TMI, Chernobyl and Fukushima all revolved around failure modes with water. The failure mode in Thorium is stopping the reactor. Will take a month or more to restart the reaction.
I also went down that rabbithole and I'm also convinced, but I'm not aware of any Thorium reactors currently in operation.
Not being a proven technology would certainly be a business risk to consider though Thorium would be an easier pill to swallow politically and socially once the population is aware of the safety benefits.
It was proven to work back in the 60's. With way worse material tech than we have today. The reinvigoration of Thorium research happened when that design was declassified and shared with some NASA folks.
The PWR tech was chosen by the US government originally because using that type of reactor supported weapons production. Many countries still want it for that reason.
Newer types aren't used yet in the US because the incredible over-regulation of existing reactors due to the fear involved has made them very un-economical to build and very difficult to get new designs approved... if there's no money in something, corporations won't do it.
What's missing is a $/kwh quote. It's important because renewables are the cheapest option on the market right now and nowhere near done dropping further in price.
If the last ten years are any indication, an order of magnitude further drop in prices before the end of the decade is not unreasonable considering mass production of batteries, solar, and wind is still ramping up and considering the enormous amount of R&D currently being spent on reducing cost and increasing efficiencies further.
In the UK it's going to be hard to compete with wind (Scotland exports more of that than its total energy consumption already) and people installing solar panels on their roof and putting batteries in their house and cars (which can power a house for several days).
No, renewable are incredibly expensive, on the contrary. Why is that? Because renewable produce when wind blows or sun shines, and not when we need power. That's the first point, we need to produce exactly as much energy as we consume, at any time. Renewables fail utterly at that, and that's why renewable energy appears so cheap when you simply consider the price at which it's sold on the markets: production occurs at random times when it isn't needed, pushing market prices down in the toilet. But that's an undesirable side-effect, and not a good thing.
Renewable produces only a fraction of their nominal output (typically 15-20%), and at random times. So typically a huge, sprawling 12km2 900MW solar plant produces a small amount of power during the year, in the best case (in the desert) 1.4TWh, while a 900MW nuclear reactor produces typically (actual numbers) 6.5TWh every year.
You also need to complement renewable production with an equivalent amount of steerable power (usually natural gas), or a large amount of batteries which multiplies the price by 5, 10 or more. So you can't consider the renewable price alone; you must take into account the price of the substitute power source, too... And its environmental price!
So renewable energy is cheap because it isn't worth much. It puts extra pressure on the grid, it forces nuclear plant to run at reduced capacity, which reduces the yield, and artificially lowers their profitability for no good reason at all.
In fact, renewables, by reducing profitability of nuclear power, has another very undesirable side-effect: making nuclear power companies cut corners on safety to save money.
Renewable power is better than coal; but it isn't, by large, as good as nuclear.
Do you have any sources for your claim of batteries multiplying the cost for renewables by "5, 10 or more" in a realistic grid scenario? Did you take into account that most energy storage is done with hydro, not batteries?
I haven't had time to carefully review the literature, but there seem to be quite a few papers that claim that storage for a 100% renewable grid would not be such a big issue.
Basically renewables produce 15-20% of the time (30 to 40% for offshore wind). Let's be generous and start with 30% of the time. You need storage to provide 70% of the demand. So for every GWh or required power, you need 700MWh of storage. The famed Tesla Mega battery is a puny 129MWh...
You must account for the fact that there could be no sun or no wind for several days, therefore be able to store incredible amounts of power, far beyond the scale we know of.Then the sheer space required for renewable power is staggering. As I said, the biggest solar plants are 600MW to 1GW and occupy tens of square km. Now imagine supplying 10s of GW this way.
As someone joked on twitter, if you covered the whole Fukushima exclusion zone (250 km2 or so)with solar panels, it wouldn't come close to the Daishi nuclear plant annual power output.
Your calculation does not make much sense (GWh isn't even a unit of power). One would have to look at actual power production and consumption curves in order to estimate the required storage capacity. Battery storage is still very exotic, pretty much the entire capacity is provided by hydro nowadays.
Current-gen solar panels average about 30 W/m² over the year in Japan's latitude and climate. So a solar park the size of the Fukushima exclusion zone would produce around 7 GW on average which is more than the combined nameplate capacity of the six reactors (5.3 GW). Supplying the entire country with electricity (~114 GW) would require about 3800 km² of solar panels, or about 1% of Japan's area.
GWh is the most relevant unit: the quantity of energy we need around the year. GW of nominal output is relevant for nuclear or fire-based power that can produce 24h/24 365d/year, but completely irrelevant for wind and solar that provide power only part time.
A 1GW solar plant outs 1GW once every sunny day, at noon. The rest of the day, it produces less, and nothing at all 12 hours a day. A 1GW solar plant roughly produces 3GWh a day. OTOH, a 1GW nuclear plant produces on a yearly average 800MWh every hour, every day and night. So at the end of the year your 1GW solar plant produced a grand total of 1100GWh, while your 1GW nuclear (or coal-fired, or gas-fired) power plant produced 7000GWh.
Plus your solar panel will provide energy mostly around noon, need it or not. While the nuclear (or gas, or coal) plant will provide power exactly when you need it.
Wind is even worse than solar, because production yield isn't even particularly predictable. So when wind blows, electricity goes into the grid without any consideration for the need for it. That's why its price drops to negative at times: because we don't need it. That's not because wind is wonderfully efficient or something. It's because building wind farms is a fantastic misallocation of resources. We're literally spending money for electricity worth almost nothing, and it won't ever recoup its costs.
That's why Germany, after having dumped 300 billions euros on wind farms in 20 years, stop building them altogether in 2018. At some point some accountant did the math.
It is completely obvious that renewables require significantly more storage capacity than other power sources. However, a lot of experts seem to believe that this is not nearly as big of an issue as you are implying.
As for wind energy, you are completely wrong. In 2019 25% of Germany's electricity was generated by wind (compared to 13% nuclear). This has significantly reduced the fossil fuel consumption and CO2 emissions.
The main reason for halting construction is that conservative politicians have passed idiotic NIMBY regulations that now make it very difficult to build new on-shore wind farms. Somehow they even managed to convince a signify fraction of the population that wind turbines are somehow dangerous (probably the same fraction that believes in electro-smog).
If Germany had invested 300 billions in nuclear instead of wind, they would have 0 emissions electric power right now.
It makes relative sense in Germany because they still rely a lot on coal for power generation. If they hadn't stopped nuclear power in its tracks, they would have achieved a much bigger reduction of their emissions, too.
If you take the example of Spain you can easily see that they fired up 1GW of gas plant for every GW of wind farm. Because some days there isn't any wind. You can't lower your emissions to zero with wind and solar.
Europe has very little room left for hydro storage. Most dams that could be built have been built long ago. That drastically limit the efficiency of more wind and more solar power. Nuclear, on the other hand, doesn't need much backup. In fact in France it needs less than 2% of backup (EDF has the know-how to modulate power very accurately to match consumption entirely on nuclear power) so it would already be possible to do 0 emission with 95% nuclear and 5% hydro and nothing else.
The other elephant in the room is that nuclear energy relies on having cheap sources of Uranium available. This works fine for the almost negligible amount of nuclear energy being generated today. Scaling this up to a significant fraction of the world energy consumption would require usage of breeder reactors which nobody has been able to get to work at the required industrial scales.
Nuclear energy is capital-intensive, but not resource intensive. The cost of the plant dwarves the cost of uranium to run it. In this way, it's similar to wind.
France had 2 working breeders reactors, Phénix and SuperPhénix, and they worked at industrial scale. They were cut off for purely political reasons.
https://en.wikipedia.org/wiki/Superph%C3%A9nix
In addition to being tedious and doing no good, such comments often end up being false, once users come along and (as they typically do) give corrective upvotes to the unfairly-downvoted comment. Unfortunately, posts like yours here don't then garbage-collect themselves—they stick around, adding noise to the thread. That's one reason we ask people not to post them.
I agree, unless there is a huge increase (2 orders of magnitude) in the ability to store energy per cubic meter per dollar then nuclear is the only path forward for bulk energy increases.
I'm running on partial "utility solar". Cost varies wildly over the year, from 1x to >3x vs regular electric; average is 1.5x ... but that's with a reliable instant backup supply for every second solar supply is insufficient vs demand.
During summer, I run my office 100% solar. It's a fun exercise, but it's deeply reliant on batteries. Storage capacity is a cliff: run out of stored power, you're done. This "brick wall effect" is grossly under-appreciated by solar proponents.
There's a funny thing about supply-and-demand: so long as supply is sufficient, costs are reasonable; when supply actually hits zero, and demand isn't zero, economic badness follows. Solar hits zero daily; batteries have a hard limit of a few hours/days.
You can solve that by installing more solar. Also, if you have an EV, it comes with a (typically 60kwh) battery. If you have the right systems installed, you could power your house from that for a few days. For reference, e.g. a tesla powerwall would come in at only 13.5 kwh (i.e. less than 25%) and you could buy a more expensive EV with 75kwh or even more. So, you could power your house using your car for a while. If you have the square meters, you could install enough solar to charge that battery capacity. It's just a matter of what still makes sense in terms of cost.
Doing stuff like this is currently expensive but will eventually drop in price to the point where it stops being a cost reducing supporting technology and can actually become your main source of energy together with maybe a more expensive option as an emergency fallback based on some grid level solution (wind, more solar, tidal power, nuclear even, gas, etc). It's just a matter of $/kwh of battery & solar panels. Both are still coming down in price. The reason people are putting this stuff in their homes is that they are already on the right side of that price compared to the grid.
They can't provide the $/kwh quote because it's not even an implemented technology yet...it's a propositional technology and they can't even provide a quote that solar/wind could smash easily in the majority of markets.
I really wanted this to be about Rolls-Royce Motor Cars Limited rather than Rolls-Royce Holdings plc. A nuclear-powered Rolls-Royce Phantom would be incredibly boss.
Actually, several American car manufactured experimented with this idea in the 50's. Most notably Ford with the Nucleon [0], but also other brands such as Packard and (I believe) Cadillac were also designing nuclear powered cars.
Eventually it was concluded that the shielding of the reactor would make a nuclear powered passenger car impractical, as it would make the vehicle too big. And of course there were the obvious safety concerns.
There is an good documentary on this subject on youtube [1].
I remember Toshiba having a decay-based laptop battery in development (at least 10 years ago, maybe 20). You wouldn't have to charge, ever. Now that would be something!
When I was five or so, I had a watch with a radium dial.
I read in a kid's science book that if you went to bed and waited until your eyes were adjusted to the dark, and then held the watch dial against your closed eyelid, you would be able to see the sparks as individual radium atoms decayed.
I did this, and it worked! I could see each spark!
Radium is primarily an alpha emitter. Alpha won't cross through the glass of the watch face, or even your skin. Maybe you were seeing the much more rare beta decays.
I doubt it was from the betas, they are all fairly low energy for typical radium. It was probably from the gammas produced by Ra-226. Ra-226 produces a 186 keV gamma in 3.3% of decays. Cherenkov radiation flashes in the eye are observed by patients undergoing clinical radiation therapy (see this interesting paper where this was explored and tested: https://www.redjournal.org/article/S0360-3016(19)33947-1/ful... ). The test done in the paper was with much higher energy gammas, but it is feasible that the 186 keV lines could cause the same effect.
I've never observed this when working with radium, but then I don't stick it in my eye.
I remember reading some witness accounts who saw nuclear explosions and they described being able to see through their eyelids (when closed) and in some cases even through their hands. Pretty wild when you think about how bright and forceful a nuclear explosion is.
Sure it would. Couple it with a lion - battery as a buffer. A tiny pellet gives 60 watts of heat. A heat cycle engine could recover maybe 10 watts of that. Put 500 pellets in there and you could get 5kw or so electric charge, all the time.
Of course the pellets are insanely toxic in just about every way.
Are you sure? If I'm reading the "Terrestrial" table on that Wikipedia page right, 5kW of electrical power would need an RTG weighing many tonnes, significantly more than a bus.
The problem with the related 25 KW of 24x7 waste heat is that'll eventually melt the very heavy several inches of lead unless very carefully cooled or distributed.
Even 20 watts will eventually melt small quantities of metal, think of every soldering iron. The authorities are not going to approve a reactor that melts its own armor if someone accidentally throws a quilt on top or equivalent.
So most designs do some kind of incredibly heavy passive cooling where the required very large surface of the armor only emits a fraction of the heat you'd see from sunlight. Now sunlight as an engineering 1 sig fig estimate is about a kilowatt per sq meter, so your 25 KW of waste heat would require some multiple of 25 square meters of surface area for passive safety cooling. A cube 5 meters on a side of solid copper and lead would have a surface area of 150 square meters, which will be warm to the touch but not as warm as it would get laying in the sun all day.
The problem with a cube of lead and copper 5 meters on a side is that is 125 cubic meters. And metal, again to one sig fig, weighs about ten thousand kilos per cubic meter. So that shielding would weigh quite a lot.
Its the old problem of the ratio of surface area to volume that comes up in so many engineering problems. Yeah I'm strong enough that I could pick up a small space craft RTG and walk around with it, for several reps at the gym anyway, but the scaling factors are such that a car charger would be immensely heavy.
I'm not sure, on a daily basis, what you'd do with 125 kilowatt-hours of car charge anyway. A gallon of gas is about 25 KW-hr (to one sig fig) so that's five gallons of gas. That would be 150 miles of driving per day, around here that would be something like 3 to 5 hours of driving per day, which sounds like a horrible way to live. Unless you're a bus driver, or delivery guy, or similar, LOL of course. Still, I don't think a five meter on a side bus or car is in our future.
> I'm not sure, on a daily basis, what you'd do with 125 kilowatt-hours of car charge anyway. A gallon of gas is about 25 KW-hr (to one sig fig) so that's five gallons of gas. That would be 150 miles of driving per day, around here that would be something like 3 to 5 hours of driving per day, which sounds like a horrible way to live
I’d have my car mine bitcoin when it’s not being used :)
I don't know about the thermal efficiency of a small nuclear reactor, but I imagine you could probably make that a closed loop system. Just as you would on an internal combustion engine, with a radiator to cool/condense the heated coolant.
Even in an ICE, you have to replenish/change the oil and cooling fluid; I'd expect a closed-loop small nuclear reactor to be similar in this regard. Still, replenishing something every couple months is much better than refueling or recharging.
"Production line mistakes may lead to generic defects that propagate throughout an entire fleet of reactors and are costly to fix," (person from University College London) warned.
Wouldn’t the knowledge gained by fixing a large one-off reactor be of limited transferable value, as compared to expensive knowledge that could be used to upgrade an existing “fleet” plus improve subsequent versions?
It also means each time you create a new reactor, you never learn from past mistakes. Thinking that mass production causes more mistakes makes no sense when you actually think about it.
Cars/guns/airplanes/even furniture all benefit from mass production to keep costs and mistakes down. There's no reason why reactors would be magically exempt from it.
I find that to be an incredibly naive statement to any student of mass production. It's not like they won't be tested on a small scale before they start pumping them out. Sure there is a minute chance but this is just fear mongering.
I'll just point out that auto makers who manufacture in the 100's of millions of units often have mandatory, stop-use, stop-sales recalls on their products 5-10 years after sale.
What you are asserting isn't the reality of mass production as experienced by industrial society so far.
1. A flaw is found in a standardized design impacting hundreds of units. Engineers identity a fix, and technicians receive training on how to apply the fix. They apply the fix to hundreds of units in a standardized and repeatable fashion. If the flaw is catastrophic, chances are this will prevent hundreds of catastrophes.
2. You have an industry producing hundreds of bespoke designs. Each time a flaw is found, engineers work to devise a solution, and technicians repair the flaw on an ad-hoc basis. The industry is competitive, so potential learning surrounding individual flaws does not propagate quickly. Each catastrophic flaw is a new catastrophe, endangering many lives.
Funny thing about auto makers manufacturing 100s of millions of units: they're still hugely profitable and despite the risk of mass recalls, bespoke designs are practically unheard of.
This is moving the goalposts quite a bit. The GP was asserting, in essence, that it was naive and fear-mongering to
say that mass-produced designs may have major flaws that need recall.
Now we get the argument that yes, that will happen but it will still be profitable to do so.
All of this contradicts 70 years of nuclear industry experience. It is an unsupported assertion, not a truth, that modular manufacturing will experience a beneficial industrial learning curve that will bend costs toward being competitive with alternatives.
Sure why not, but let us not get carried away and think THIS is the solution to global warming. It is unproven, it will be too late, and not scale up quick enough.
By unproven I mean that the economics of it is unproven. We have no proof that SMR will be cheaper than large reactors. The article points this out as well.
By late I mean at current fast pace of development a lot of will happen with wind and solar in the 10 years it will take for this to get online.
Wind and Solar need battery backup. There is no grid-scale battery technology today, nor is one forthcoming in any sort of reasonable term (on the order of the next one or two decades). Care to rethink your standard on what is proven and unproven?
Having worked in the industry, I like seeing companies pledging to get involved with nuclear power.
However, there is a dearth of people experienced with the design and development of new reactor technology. The development pathway of building up the supply lines, understanding the complexities of the various systems, etc. is long.
I laughed at first many years ago when I saw some new reactor designs taking 10years but it turns out in a lot of cases that's optimistic.
It's a shame, because development in the mid 20th century was a LOT faster. And the people involved moved from one new reactor design to the next, bringing valuable experience.
Much of the work force entering this industry are entirely green and need an understanding of existing reactor technologies and why some designs were made a certain way before trying to reinvent them. Looking at the development in Russia and India could be very helpful.
Personally I'm all for having mini reactors if it means we can generate reliable energy with minimal pollution. Anything that helps reduce emissions should be considered.
But this would need to be a fail save type of reactor and with a clear plan(s) all the way down to how the waste is handled till it's no longer active.
TerraPower [1] seem to have a good alternative with its Traveling Wave Reactor. It can be run on nuclear waste we already have and provide energy to 10bln people with US per capita power consumption for basically eternity.
These are well-known properties of any fast reactor, like Terrapower's.
Only about one percent of high-level nuclear waste is fission products, the broken-apart atoms left after fission. The rest is U238, unused U235, plutonium, and other transuranics produced by absorbing neutrons without fissioning. Fast reactors can fission all of these. That's why they can run on nuclear waste.
For the same reason, they can get over a hundred times as much energy from the same amount of uranium ore. Conventional reactors can only use the U235, which is 0.7% of natural uranium.
That's a great start, but with such efficient use, it's practical to get the uranium from seawater. We can do that now at five times the cost of mining; if we only need 1% as much uranium, then total fuel cost would be 5% as much as nuclear reactors spend on fuel today. Fast reactors fueled from seawater would last for millions of years.
"TerraPower planned to build a 600 MWe demonstration Plant, the TWR-P, by 2018–2022 followed by larger commercial plants of 1150 MWe in the late 2020s.[15] However, in January 2019 it was announced that the project had been abandoned due to technology transfer limitations placed by the Trump administration."
I'd much rather nuclear production operations be handled by larger plants, and storage be handled by the customer. Storage efficiency is the current weak point, not production efficiency.
I'm for them if they make economic sense and also can't go critical, as in you can more or less hit a switch and they go dead and criticality isn't a possibility. There have supposedly been designs that guarantee that, but I saw nothing in the article about it.
I long said that "mini" reactors don't make sense when 1.7GW reactor and a mini one both have dramatic minimal cost for a single unit, initial startup cost, and operational costs.
Only if mini-reactor gets cheaper than the coal plant, can they get competitive.
The natural reserve of radioactive material, like that of fossil fuel, is finite. We cannot count on it in the long run. On the other hand, we are awash with the huge amounts of the free solar energy. We just need better batteries, which is a purely technical problem.
Uranium seawater extraction makes fission fuel effectively limitless [1], on the order of billions of years of supply. If you want to get pedantic, it is limited but so is the hydrogen in the sun.
I've been thinking about this for a bit and had a similar model in mind. You build small nuclear reactors in a containers - similar how Google builds their DC. You find an abandoned coal mine shaft and put them deep in the ground. The cost optimizations come from:
- economies of scale
- no decommission costs - you just leave them to decay.
I'm not sure how do you mean. You get economy of scale by saving in costs gained by an increased level of production.
It's cheaper to create 100 widgets for 100 small reactors in 10 locations, then for 10 large reactors.
I disagree. If one generator fails, it's not a big problem. If you have a failure in a large reactor it's a biggie, so you have to use very expensive materials and design to avoid failure.
I wish them luck, but the projected price of £70/MW hr sounds very expensive. I gather the retail price in the UK is about double that. That must put them right on the edge of profitability now, and the things have a design life of 60 years. If they have to hit that 60 life to get that £70/MW hr they must know they are already screwed, surely?
OP is probably asking some variation of why not build them on a canal rather than literally right on the coast, Fukishima-style.
Part of the problem is the components are very large and very heavy even for industrial projects and coasts have barges and seaports.
Some parts are too heavy even for railroad shipment, so its barge and crane or expensively fabricate on site, or expensively assemble on site.
Its quite a logistical headache to build a large nuclear reactor. Yes, submarine scale plants are petite, but GW-class baseline generators have very large and heavy parts.
WRT to Fukishima, its probably cheaper and more efficient from an engineering perspective to build a massive seawall than an unblockable long canal. The effort and concrete required to build an unblockable kilometer long canal that would survive a 30m tsunami would probably survive a 100m tsunami if put into a seawall instead. Or more likely, management would only permit a 10m rated seawall (or canal) "because that'll never happen" and we all know how that turned out.
Rivers tend to have different flow rates over the course of the year. If there is a heat wave the plant may not have enough water for cooling. The ocean is more dependable.
The problem is not so much the flow rate as it is the river's temperature. You can't raise that too much without harming the river's ecosystem. This is what limits several of France's nuclear power stations during summer.
Here in Canada in my region the nuclear power plant is located on a bay in New Brunswick. The bay has the highest difference in tides in the world at high tide the water is 16m (53 feet) higher than low tide.
But in Ontario where the biggest nuclear power plants are located. Two are next to Lake Ontario, and one is by Lake Huron.
That sounds like a very bad idea. Although I am confident that the tech has evolved since Chernobyl, terrorism has evolved as well. It just takes a bunch of assholes to destroy a small town.
And yes, I think they will be pretty well guarded. But if you have lots of them some of them will have security flaws.
I stopped worrying about terrorism when I realized that they could easily take out most of the European power grid with a few well-placed bombs on the central very-high-voltage power lines. Yet for some reason, they don't.
If terrorists were really hellbent on killing as many enemy civilians as possible, the world would look very different. What actually seems to be happening is that they use the smallest possible intervention that causes sufficient fear (and, conversely, sufficient support on their home turf). So I don't think they'll be blowing up nuclear reactors anytime soon. That would be unnecessary overkill (literally).
Really terrorism is like the Spanish Flu - while not good it is the knee-jerk excessive reaction that is the true threat.
I suspect it is a matter of ideology and psychology. Just emptying an AR-15 clip into substations throughout a country isn't "exciting" or "sexy" even if it would be very effective saboage. It is boring and doesn't validate their ego or whatever statement they want to make. It would be petty vandalism and not even inspiring to those already radicalized to their cause. Plus while causing damage and disruption it isn't exactly terrifying - the aftermath involves selling bonds and overtime for repair and manufacturing. Even if it kills it is far less traumatic than direct violence.
Atomwaffen are essentially the only group I know of who would actually have something like that as a goal instead of a potential method. They are essentially an example of the theory of fascism as a death cult insanely trying to control their own mortality it by inflicting death upon the other. Thankfully they are fundamentally incompetent losers.
Interesting idea. I'd say they are just not intelligent enough to really figure out how to kill as many as possible. If they where intelligent, they would understand that terrorism rarely works in achieving its stated goals and do something more likely to work out in their favor.
You can't just walk into a nuclear reactor and take away bomb grade material. Even if you run a reactor, you can't get bomb grade material out of it if you allow regulators in.
Dirty bombs are also far less dangerous as they are made out to be.
Assholes can already destroy small and even large towns by conventional means. I'm sure you yourself can come up with a dozen ways of killing everyone in the town you are in via conventional explosives, poisons, toxins etc.
This is both a flawed argument, and a point addressed directly by the article – with the proposal being to place these reactors in existing nuclear sites with security strategies in place.
"Survivorship bias or survival bias is the logical error of concentrating on the people or things that made it past some selection process and overlooking those that did not, typically because of their lack of visibility"
Can you point me to an example that is being overlooked?
The probabilities of a successful terrorist attack on a nuclear plant are very thin given the extreme security measures around them, and that's a fact.
Thinking that, despite a basic probability being extremely small, "it's still possible", and thus assigning the event a larger probability than it naturally has, now that's an error.
You expect a technology to be made to work first, then to be made to work profitably, and only then commodified, miniaturised, etc.
Nuclear works, but no one has made it work profitably. There is no reason to think shrinking it will make more profitable.
Quite the opposite in fact: nuclear has big fixed costs. That's why you usually see multiple reactors build close together on single sites. The more profitable (less loss making really) sites are the biggest ones.
I believe you have not disagreed with me, but you have introduced additional intermediate stages and conditions with which I disagree.
I disagree that it is necessary for technology to be made profitable before becoming commoditised. I would argue that some technologies only become profitable when they are commoditised.
There is no reason to think shrinking it will make more profitable.
I cannot see into the minds of Rolls Royce, but I would hazard that they think it can be done. They suggest that if they can export this technology overseas as well, they can turn a profit on it. While I don't know if they're correct, they clearly believe it's possible.
I think we assume things shrink and commoditise because technology we use daily has often done that (say home computer parts). But other technologies, especially industrial ones, have not.
Cars for instance are bigger and offer more diversity and less commoditisation than ever before.
Similarly, we've had coal power plants for a long long time. And they are larger than ever. Do you know anyone with a mini coal plant? Neither do i.
Nuclear specifically shows no likely hood of shrinking or commoditising in my opinion. Large fixed costs make shrinking hard. A wide diversity of requirements from customers and interconnectedness makes commoditisation hard.
Maybe rolls toyce have solved 2 problems when no one else has solved 1, but I'm skeptical...
He's also failing to identify hidden costs like destroying the global climate that we're most likely going to absorb at the government level despite what greedy lasseiz-fare capitalists want to admit
Betting on a world wide, full funded program to reduce carbon is... Courageous. Betting it will use nuclear and small nuclear and rr's small nuclear and that they will get it ready on schedule and deliver it in time to make any difference? Let me sell you some magic beans I have.
Nuclear reactors' biggest drawback is the cost and danger of dismantling and disposing of them after their working life is finished. Up till then they are clean and cost-effective.
Why does no one ever factor in the cost of dismantling other major projects, like the Hoover Dam, or the Three Gorges Dam? ...or thee solar plants being built, or the wind turbines?
Why does nuclear power have this enormous additional requirement that literally no other power source seems to have to account for?
> Why does nuclear power have this enormous additional requirement that literally no other power source seems to have to account for?
Because no other power source has the inner components become radioactive during normal operation. Normally these radioactive parts are fully contained, but that's no longer the case once it's dismantled. That makes dismantling a nuclear power plant not only much more expensive, but also a requirement (you can't just abandon it, otherwise it will gradually break down and lose the containment).
The huge cost leads to mismanagement. Obsolete or unsafe nuclear plants keep running even when they shouldn't. Those plants should have been replaced with new designs immediately but it is cost prohibitive and that is what lead to accidents like Chernobyl.
If you can't afford to shut it down then don't build expensive big designs. Build smaller ones that you can actually afford to take offline.
Chernobyl's design was amazingly primitive compared to the vast majority reactors outside of the former USSR. Any reactors that unsafe in any western nation would be immediately shut down.
We shouldn't be using nuclear energy for power. Solar, wind and batteries will solve that.
What we should be doing is using nuclear energy for HEAT. We need a LOT of heat for a variety of chemical and industrial processes, not to mention heating of homes in colder climates. For these uses nuclear fission has massive potential.
That’s one of those ideas that seems great on the surface but is much more expensive to implement than you might think. Moving heat around involves significant energy loss over long distances. That’s offset with combined heat and power generation which is very popular in Russia etc, but on it’s own it’s a tougher sell.
Why not both ?
With cogeneration you raise the efficiency from only electric generation from 40% to 80% with heat, but in summer you can still generate electricity only.
And SMR are better positioned than classic reactor because the smaller size and inherent passive security from surface to volume problem (heat raise with volume in power of 3 and cooling only with surface in power of 2) make them easier to build them near populated cities.
I should have been more nuanced: we shouldn't be building out the current design of fission power in order to decarbonize our electricity production. The biggest reason is that it just takes far too long to build: we're talking 15-20 years. Wind and solar can scale far more quickly than that.
Closing existing nuclear power plants like they're doing in Germany is absolute lunacy.
For heat that timeline is acceptable given that it can be "decarbonized" by switching to burning biomass to cover the transition.
If we could build safe nuclear electric plants within a reasonable timeframe and cost then we should absolutely do it (this would make so much sense in huge countries like the US or China because you can have the reactors located well away from large populations).
I wonder, though, how much we could bring the price of nuclear down if we were to actively develop the technology, rather than letting it languish. I also wonder how much toxic waste batteries will produce vs nuclear. Do we have clear numbers on that?
Pretty sure that's not true if you account for total costs. (Including the energy and materials budget for making solar panels and batteries at scale.)
Yes I try to bring it up to my "only natural sources!" friends but they won't hear it. We need at least a 100 fold improvement in energy storage to make it practical and provide "base load" on par with what we have now with nuclear and fossil fuels.
Solar + batteries averaged over the 24 hour demand curve are cheaper than current generation nuclear power plant operating 24/7/365. The issue is turning off nuclear power plants don’t significantly reduce costs so their competing head to head vs x hours @ 2c/kWh + y hours @ ~20c/kWh.
At first glance it seems to slightly favor nuclear when it can operate 24/7, but nuclear power plants have a very long lifecycle 40+ years and projected battery costs for future replacements are lower than today’s values. Worse nuclear has a much longer lead time so you already need to compare battery prices ~5 years from now when you’re looking at building a new nuclear power plant.
PS: Note grid batteries are significantly cheaper when they can share inverters with solar generation and be directly charged via DC. Standalone prices are higher. Also, fast discharge peaking power reduces battery lifetimes over longer discharge nighttime useage.
Batteries inherently allow for peaking power, it’s nuclear that has issues with fluctuations in demand. So, by average I mean how much are you able to use directly vs how much do you need to store in batteries.
To be more clear only using solar with battery backup you can provide 24/7 power that covers demand but at lower costs per kWh than current nuclear numbers operating 24/7 and vastly lower than trying to reach 100% nuclear generation in almost every country.
Nope, ~60% of France’s power is provided by nuclear while 71.6% of it’s power production is nuclear. They export 86.3 TWh almost entirely nuclear power and import 26.1 TWh almost entirely non nuclear power (2016).
Even then their nuclear capacity factor is 77% where it’s generally 90+% in the US because frequently nobody want’s to import electricity from France. So, France’s high percentage of nuclear power results roughly 17% higher costs per kWh for nuclear power due to oversupply. Or more depending on what it exports at.
Further, the more you increase nuclear capacity globally, the less anyone is going to be importing it at night and weekends when everyone is over supplied.
I don't know about a lot. I live in Tromsø, Norway (69N). There aren't many populated areas this far north. And even here, solar is pretty decent for a lot of uses (Esp in combination with insulation to reduce the need for heating).
We also have quite a lot of precipitation - but again, solar isn't hopeless even here.
As for seasonal storage, I suspect some kind of kinetic/potential energy storage might be better than batteries? Like lifting a weight, or pumping a liquid up hill/up a tower?
Yes, a mix of technologies. But there's more to it than that. We're going to need increasing amounts of power in the future to explore the solar system and to solve all kinds of existential problems. Energy usage is not inherently evil provided we learn how to do it safely, which is an ongoing process.