5 ms·

(Background: I am an engineer that has spent most of their career in energy - fossil fuels, nuclear and renewables - the whole shebang. I care deeply about clim

by orson2077 2y ago

(Background: I am an engineer that has spent most of their career in energy - fossil fuels, nuclear and renewables - the whole shebang. I care deeply about climate change, and recognise the non-negotiable need of humanity for ever-increasing amounts of energy.) Take it from me - I would absolutely love renewables to displace all other forms of energy. However, from an engineering point of view, I know this is simply not feasible without some strong developments in energy storage and distribution. With today's technology, nuclear is the only real threat to the hegemony of fossil fuels, and the only practical hope of reaching climate goals (let alone long-term energy security goals). I think the fossil fuel industry knows this, and so frame their anti-nuclear rhetoric as a renewables-vs-nuclear debate, backing renewables. If I wore a tinfoil hat, I would say that a lot of the support for renewables (and hostility towards nuclear) was encouraged/stoked/funded by the fossil fuel lobby. There's an interesting blog that catalogues cases where the fossil fuel industry has scratched on nuclear. If you're interested, I'll link it below. https://atomicinsights.com/smoking-gun/ https://atomicinsights.com/smoking-gun/ As for SMRs specifically, I think they have a lot of legs. Reactors that can be assembled (and serviced) in factories would go a long way to lowering the overall cost of the reactors (even accounting for the reduced energy-cost density of the individual reactors). Questions welcome. I created an account just to give my two cents on a field to which I have dedicated my career.

julkali 2y ago
How would you then argue against the points listed in the article?
- orson2077 2y ago
  Regarding their main points: - Too expensive. Nuclear power plants usually operate for 40-80 years, making their ROI after the 20 year mark (greatly varies). The report's choice of "10-15 years" for a return on investment is suspicious, as it corresponds more to the life of PVs and wind turbines. - Too slow. The first instances of a new design always take longer than the mass production instances. It's madness to compare prospective factory-manufacturable reactors to the behemouth reactors we are used to today. (Also, from memory, I think Japan once made one of those behemoth reactors in 22 months... delays are often not for technical reasons). - Too risky. Without storage and/or distribution solutions, renewables will inevitably depend on fossil fuels; this applies both to service economies and manufacturing economies. The difference is that nuclear captures is externalities, unlike fossil fuels. - A bad fit. I actually agree with this one in some cases. For example, Australia has abundant land and great weather; they could probably get by with pure renewables. However, countries like Germany (which has so-so weather and some heavy industry) would be hard-pressed to do the same. They could achieve 100% renewables by giving up certain industries, but I don't think that's reasonable to ask. - The Boeing Problem Boeing's fall from grace has everything to do with perverse incentives and regulatory failure. If the public is crucifying them for dodgy planes, I imagine they'd do even worse for making dodgy reactors. Regulation is a must for nuclear, and never has anyone serious thought otherwise.
  - credit_guy 2y ago
    I love nuclear stuff, and I agree with you that SMR have a very good chance to become cheaper. I think we, the society, should invest in nuclear (specific nuclear fission) because of the immense energy density of uranium and thorium. Yet, I think reaching our climate goals is entirely doable without nuclear. Why? Net zero does not mean zero emissions. It means emissions equal to sinks. Right now in the US all the emissions coming from natural gas power plants are equal to all the sinks (generally forests) [1]. When I tell people that they are surprised. Here's the numbers: electricity contributes 25% to the emissions, and natgas power plants generate 45% of the emissions associated with power generation [2]. So 11.25% of emissions come from these power plants. The greenhouse gas sinks for the US are at 13%. So, if we ditch all the coal power plants (which is happening right now, at high speed) and we build a lot of solar and wind, and keep all the current natural gas power plants as peakers, then we will be well below net zero. [1] https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas... [2] https://www.eia.gov/tools/faqs/faq.php?id=77&t=11 https://www.eia.gov/tools/faqs/faq.php?id=77&t=11
  - rstuart4133 2y ago
    > For example, Australia has abundant land and great weather; they could probably get by with pure renewables. Maybe we (Australians) can do 100% renewable. We will see. But even if we never do (and it's not entirely clear how it's possible), it's hard to see a place for nuclear here. Since you say you follow this closely, you are probably aware one Australia state is at 70% renewables. That's 70% average, over a year. Unlike other places you hear about with a lot of renewables, South Australia has no hydro. Like the rest of Australia SA is pretty flat, so it has no pumped storage either. In fact there is nothing special about SA at all, other than it has no coal or gas, and is at least 500km from anywhere else of note so transmission lines cost a small fortune. It's not an ideal place for renewables, but beggars can't be choosers. I'd love to say SA hitting 70% was a master stroke of forward planning. It was anything but. You will hear some politicians claim the did it for climate change. Maybe it was, but what they did happened to coincide with taking cheapest option on the table at the time, over and over again. Solar and wind are damned cheap when they are only contributing 10%. Getting to 70% is more expensive, but they already had the natural gas peakers so at each step the options on the table were to import more natural gas, or put up a wind turbine and use less gas. Each % reduction gets asymptomatically more expensive of course. Over provisioning helps, but typically solar drives the price negative during the day now. They claim they will get to 100% in 2027, but without storage I don't have a clue how that's possible without using the transmissions lines to states with coal generators and some creating accounting. It's possible the current 70% made the grid a unstable. It's hard to know. They did loose power for days, but the proximate causes were some transmission lines were blown over in the worst storm in decades, inter-state interconnects were down for maintenance, and wind turbines tripping out because of the spikes created by the first two. I'd love to say that had been anticipated and they were the victim of delays in building storage, but storage was deemed to be a money losing proposition. Hell, I'd even like to say the engineers stood up and said "we can fix this with a battery", but that didn't happen either. What actually happen is there was a political shit storm over whether the outage was caused by renewables, and the SA government found itself under an enormous amount of pressure to announce a fix. Elon, the masterful dick waving salesman that he is, proclaimed he could fix the issue by installing the world's biggest battery in 100 days - or it was free. It made headlines around the country, and they took him up on it despite the fact that it cost a small fortune and everyone knew it would lose money. This is how the decision making process has always been. A complete cluster created by special interest groups fighting over their preferred way. It seems everyone hopes to win the fight by yelling at each other, including engineers like yourself. In the end the pollies throw their hands in despair and choosing the easiest option at the time. Everyone, and I do mean everyone including the engineers was wrong about that battery. It made money from the day it was installed. Turns out when a coal fired generator trips out and removes megawatts from the system in a single 50Hz cycle, to the computer controlling that battery that 20 milliseconds looks like an eternity. It can react in microseconds and dump compensating power into the system long enough for a peaker to fire up. And charge a small fortune for doing it. Apparently no one foresaw this, and so no serious grid scale batteries were added. Now everyone has seen they make money new battery installations are springing up like weeds all around the country. Again I'd love to say they are doing it for the climate or for grid stability but no, they are doing it to get on the gravy train. The way we are going about this transition is nothing if not consistent. Predicting what the end game looks like seems like fools errand to me. 100% renewable seems dubious. But they are at 70% now, so somewhere above 80% seems reasonable by 2030. Maybe they will start building pumped storage by 2050 - stranger things have happened. But using an SMR to fill that 20% gap - that's beyond strange. It's a shitty 20%. You are off most of the time. You have to go from 0 to full peak within an hour or so when the wind stops blowing and the sun ain't shining, and then from full peak back to 0 on a dime. Nuclear might be good at a lot of things, but the one thing it's absolutely hopeless at is load following. Japan having the worlds highest percentage of nuclear is also why Japan has the most pumped storage per unit generation. The economics aside, nuclear would require a lot of forward thinking and commitment. Clearly when it comes to power generation that hasn't been Australia's strong suite. I don't see it but maybe there is some place in the world that is different. (Who could possibly have thought buying gas from Russia was a good idea?) China would be a good candidate I guess. China does have a 30 year plan for building nuclear. But with the price plummeting on renewables it looks like they've now abandoned it in all but name as they are adding more renewables each year than the nuclear plan called for over it's entire lifetime.
hamilyon2 2y ago
What do you think about power plants apparently needing a lot of fresh, relatively cold, relatively clean water? A precious resource in future.
- orson2077 2y ago
  Modern powerplants are closed-loop and do not consume water, although they may dump heat into it. Water is not consumed (in any great quantity) or contaminated, except that which is recirculated inside.
  - pfdietz 2y ago
    > although maybe they may dump heat into it. Dumping heat into water is how they consume water. They either evaporate the water directly, or they heat a river, causing it to evaporate more later.
- cyberax 2y ago
  Power plants can be designed to work in hot deserts. The largest nuclear power plant in the US (Palo Verde in AZ) gets cooled by evaporating treated wastewater. French nuclear power plants were just not designed for droughts.
- schmidtleonard 2y ago
  They need heat sinks, not fresh + cold + clean water. Even the heat sinks are only really necessary due to concentration of generating capacity rather than the amount of generating capacity. For example, photovoltaic usually has thermodynamic efficiency around 20% while steam plants (nuclear, fossil, geothermal, etc) are usually around 33%: solar panels will release considerably more heat into the environment per unit of energy generated, but since it's spread out nobody cares. Small Modular Reactors are a big step in the "spreading it out until it's easy to get rid of" direction. What does need (and not just "need" but actually "consume") fresh, cold, clean water (and dry air) is swamp cooling, which for some reason seems to do the rounds as as environmental silver bullet every few years. But that's a different rant.
DarkNova6 2y ago
Any design which you are favoring in general? From an outsider's perspective, SFR based designs look like hellish machines which are begging to burst up in flames. In contrast, I love the design of the properties LFR, except the material problems of high temperature lead. And in German circles the Dual Fluid reactor gets a lot of buzz as well.
- orson2077 2y ago
  I have a tremendous love for the fast breeder reactors (particularly Superphénix); that we had that technology so long ago astounds me. However, I recognise that they are technically challenging. From a modern, pragmatic point of view, I'm very partial to SMRs and AMRs (advanced modular reactors). They are also easier to implement on non-technical grounds (e.g. site permissions).
  - pfdietz 2y ago
    France has basically abandoned their fast breeder effort. They defunded the next fast reactor.
- cyberax 2y ago
  Lead-fueled reactors are definitely an interesting area, and Russia is in the process of building a 300MWe demonstrator (BREST-300-OD). That being said, they have their own share of problems. Startup and maintenance is going to be a beast, slowly heating up the reactor to working temperatures is going to take almost a year. The other problem is that it's still not clear if corrosion problems can be solved satisfactorily. Reactor construction steel is stainless because it has a protective film of oxides on its surface, and lead gradually rubs it away (and then rubs away the construction steel, at a much faster pace). BREST-OD reactor designers spent a decade perfecting a system that is supposed to manage the amount of dissolved oxygen in lead, but it needs to be tested in the actual reactor conditions. With all its crazy temperature gradients and flows. Another interesting area is light-water cooled breeders. Such reactors _are_ possible, but just marginally. And they have the nasty positive void coefficient, but it looks like it should be possible to compensate for it.
matthewdgreen 2y ago
Sure. I have questions: 1. We need to be at net-zero by 2050 (and realistically based on the fast-moving temperature increases happening around us) maybe decades before that. SMRs haven’t been practically deployed yet, and don’t seem likely to be even minimally deployed until at least the 2030s if ever. How are we going to displace all forms of fossil fuel on the miniscule runway we’ll have left? 2. We have to deploy vast numbers of SMRs to the entire planet, including places with much worse security guarantees than first world nations. How do we propose to secure the huge amounts of fissile material and waste these reactors will use/produce. I’m not worried about fission bombs necessarily, but I am worried about pollution and dirty bombs. 3. “Without major improvements in storage” is doing a lot of heavy lifting. We’re seeing massive declines in storage prices and huge increases in production. If storage follows a curve similar to Solar PV, a huge fraction of the profitable applications for nuclear will be gone to cheap renewables and storage. Even if it doesn’t, renewables and today’s batteries are already driving fossil sources off the grid. How do we pay for a technology that will only have a few use-cases left after all the low hanging fruit has been eaten? PS The last question is not a troll. It’s very obvious that renewables are going to generate something like 80-90% of our power. I’m open to the possibility that nuclear could make up the remaining 10-20%. But the economics of that chunk will be messy, since SMRs will have to compete with dirt cheap electricity when the sun/wind are available (even if storage stays expensive.) I highly doubt that SMRs are going to outcompete Solar PV when the sun is shining. What do the economics of the mostly-renewables-and-storage-with-SMRs-as-backstop world look like?
- ants_everywhere 2y ago
  > I’m not worried about fission bombs necessarily, but I am worried about pollution and dirty bombs. Realistically, the nuclear industry will just decide in a few years that it's too hard to properly dispose of waste and they'll dump it unsafely somewhere that it harms humans. This will be easier to do in countries that already have large hazardous waste problems. In the US it will require a great deal of lobbying, but at least 1/3 of the population will support it if they think it makes the economy stronger and sticks it to their enemies.
- h0l0cube 2y ago
  > We’re seeing massive declines in storage prices and huge increases in production. Just in the last 9 months I’ve seen LFP prismatic cells from China drop 50% in price (and about 10% in pack volume). Add to that the variety of grid scale storage tech emerging out of R&D into the real world, by the time a nuclear reactor built today is coming online, it will be obsolete. More importantly, as renewables reach overcapacity, we need fast dispatch ‘peaker’ plants, not baseload, which makes gas a better transitional power source than nuclear, and batteries the end game
monkaiju 2y ago
Wild to me how someone can open with something as ridiculous as recognising "the non-negotiable need of humanity for ever-increasing amounts of energy" and then be taken seriously. Its not only negotiable, its a requirement that we do not maintain such consumptive expansion.
- orson2077 2y ago
  Hey, I'd love to live like the elves in Lord of the Rings, but it's not gonna happen. The first world offloads its manufacturing burden to the third world, then criticises them for polluting the atmosphere with fossil fuels. Also, the majority of the world do not enjoy the standard of living we do in the first world, and they will want to. Like it or not, energy demand will increase. It is non-negotiable unless you think there'll be some great die-off.
fragmede 2y ago
What would you do with a billion dollars?
fragmede 2y ago
Do you think we'll crack fusion?
olau 2y ago
The "strong developments" in the grid and storage you speak of is peanuts compared to trying to make nuclear cheap. Big, powerful companies with brilliant physicists and engineers have worked on it for decades. Lots of prototypes, lots of designs that didn't work out in practice. Today, we don't actually need new tech for distribution and storage, although new tech is being developed and helps. Chemical batteries for small-scale time-shifting and hydro plus gas plus biomass for larger scale. In 20 years, the youth of today will have to decide if they want to get rid of the last fossil gas - long term storage is on the order of 10% of total energy needed in the studies I've seen. It can be substituted with gas made from biomass. In fact, it's already happening in the country I'm in. Distribution is just building more connections. I think working on improving the cost of underground connections would help, but it's doable with today's tech, and it is happening. If anyone's interested, there are plenty of academic papers discussing this, and also a bunch of more accessible articles by this dude: https://cleantechnica.com/author/mikebarnard/ https://cleantechnica.com/author/mikebarnard/ For instance, here's a recent one https://cleantechnica.com/2024/04/11/the-short-list-of-climate-actions-that-will-work-india-edition-transcript/ https://cleantechnica.com/2024/04/11/the-short-list-of-clima... And yes, it's a complex topic. And yes, there will be some pain points along the way - energy is important to modern society, and it is a large transition that will years to unfold.
kkfx 2y ago
A personal take: the point for SMRs it's not costs nor potential mass-production and so on but a simple thing, they can be moved. We are in a changing world, we know, at least some of us know, we don't really know the future, but we know many will migrate and this means wars but also the inability to made fixed land infra, like roads, rails, electricity grid etc. Long story short: we need to have electricity "with us, moving" and p.v.+storage can works for a large slice of inhabited land but not enough, so we need something similarly easy to locally deploy. What do you think about that?
_ph_ 2y ago
I am sorry, but I think most of what you wrote is plain wrong. Yes, I agree, the total amount of energy used by humanity is going up. Especially as there are so many nations on earth which so far only use little energy. On the other side, it will somewhat go down from the current state for many leading industrial nations. However, the bigger change will be how we use energy. In the last century the whole grid was optimized for a mostly constant load because that is what the then-used technologies, nuclear and coal, could do best. There were even big incentives given to customers to have a mostly constant draw of power. But now this changes. We have energy sources which produce electricity very cheaply, but not at a constant rate. And the experience shows, if the end customer is charged by availability, the consumption patterns change to optimize costs. This will be a big factor in the rollout of renewables. As the article shows, cost-wise the SMRs can't compete. Renewables are getting ridiculously cheap. And even more: every one can set up renewables. You can go and buy yourself solar panels and put them onto your house or into your garden. The same advantage applies to industry-scale deployments. Same with wind. The only disadvantage with wind are the permits, as wind generators are quite big. But those are still trivial compared to nuclear, as the wind generators are not dangerous in any other sense. Because they are cheap, the electricity markes are going to be flooded by renewables. That is basic economy. So the question is, how can we complement renewables best, especially to cover the gaps in their production. It won't be one thing, but a combination of several options as they are non-exclusive. Local conditions will put a stronger emphasis on one technology vs. the other. But one thing is clear: it won't be nuclear. Because even in the most optimistic szenarios, nuclear is a bad counterpart, as it is not good at providing varying output. Even if that is technically possible (usually it is not), economically it doesn't make any sense at all. For a transition time, it will be gas, as gas plants are relatively cheap and fast. Gas is expensive, so running the plants infrequently makes sense. The gas can later be provided out of renewable production. Most likely, battery storage will kill that too. But if not, the gas power plants can still be used. And in all of that I haven't even touched the operational safety, nuclear waste, and of course giving nuclear technology to countries we don't trust.

julkali 2y ago

How would you then argue against the points listed in the article?

orson2077 2y ago
Regarding their main points: - Too expensive. Nuclear power plants usually operate for 40-80 years, making their ROI after the 20 year mark (greatly varies). The report's choice of "10-15 years" for a return on investment is suspicious, as it corresponds more to the life of PVs and wind turbines. - Too slow. The first instances of a new design always take longer than the mass production instances. It's madness to compare prospective factory-manufacturable reactors to the behemouth reactors we are used to today. (Also, from memory, I think Japan once made one of those behemoth reactors in 22 months... delays are often not for technical reasons). - Too risky. Without storage and/or distribution solutions, renewables will inevitably depend on fossil fuels; this applies both to service economies and manufacturing economies. The difference is that nuclear captures is externalities, unlike fossil fuels. - A bad fit. I actually agree with this one in some cases. For example, Australia has abundant land and great weather; they could probably get by with pure renewables. However, countries like Germany (which has so-so weather and some heavy industry) would be hard-pressed to do the same. They could achieve 100% renewables by giving up certain industries, but I don't think that's reasonable to ask. - The Boeing Problem Boeing's fall from grace has everything to do with perverse incentives and regulatory failure. If the public is crucifying them for dodgy planes, I imagine they'd do even worse for making dodgy reactors. Regulation is a must for nuclear, and never has anyone serious thought otherwise.
- credit_guy 2y ago
  I love nuclear stuff, and I agree with you that SMR have a very good chance to become cheaper. I think we, the society, should invest in nuclear (specific nuclear fission) because of the immense energy density of uranium and thorium. Yet, I think reaching our climate goals is entirely doable without nuclear. Why? Net zero does not mean zero emissions. It means emissions equal to sinks. Right now in the US all the emissions coming from natural gas power plants are equal to all the sinks (generally forests) [1]. When I tell people that they are surprised. Here's the numbers: electricity contributes 25% to the emissions, and natgas power plants generate 45% of the emissions associated with power generation [2]. So 11.25% of emissions come from these power plants. The greenhouse gas sinks for the US are at 13%. So, if we ditch all the coal power plants (which is happening right now, at high speed) and we build a lot of solar and wind, and keep all the current natural gas power plants as peakers, then we will be well below net zero. [1] https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas... [2] https://www.eia.gov/tools/faqs/faq.php?id=77&t=11 https://www.eia.gov/tools/faqs/faq.php?id=77&t=11
- rstuart4133 2y ago
  > For example, Australia has abundant land and great weather; they could probably get by with pure renewables. Maybe we (Australians) can do 100% renewable. We will see. But even if we never do (and it's not entirely clear how it's possible), it's hard to see a place for nuclear here. Since you say you follow this closely, you are probably aware one Australia state is at 70% renewables. That's 70% average, over a year. Unlike other places you hear about with a lot of renewables, South Australia has no hydro. Like the rest of Australia SA is pretty flat, so it has no pumped storage either. In fact there is nothing special about SA at all, other than it has no coal or gas, and is at least 500km from anywhere else of note so transmission lines cost a small fortune. It's not an ideal place for renewables, but beggars can't be choosers. I'd love to say SA hitting 70% was a master stroke of forward planning. It was anything but. You will hear some politicians claim the did it for climate change. Maybe it was, but what they did happened to coincide with taking cheapest option on the table at the time, over and over again. Solar and wind are damned cheap when they are only contributing 10%. Getting to 70% is more expensive, but they already had the natural gas peakers so at each step the options on the table were to import more natural gas, or put up a wind turbine and use less gas. Each % reduction gets asymptomatically more expensive of course. Over provisioning helps, but typically solar drives the price negative during the day now. They claim they will get to 100% in 2027, but without storage I don't have a clue how that's possible without using the transmissions lines to states with coal generators and some creating accounting. It's possible the current 70% made the grid a unstable. It's hard to know. They did loose power for days, but the proximate causes were some transmission lines were blown over in the worst storm in decades, inter-state interconnects were down for maintenance, and wind turbines tripping out because of the spikes created by the first two. I'd love to say that had been anticipated and they were the victim of delays in building storage, but storage was deemed to be a money losing proposition. Hell, I'd even like to say the engineers stood up and said "we can fix this with a battery", but that didn't happen either. What actually happen is there was a political shit storm over whether the outage was caused by renewables, and the SA government found itself under an enormous amount of pressure to announce a fix. Elon, the masterful dick waving salesman that he is, proclaimed he could fix the issue by installing the world's biggest battery in 100 days - or it was free. It made headlines around the country, and they took him up on it despite the fact that it cost a small fortune and everyone knew it would lose money. This is how the decision making process has always been. A complete cluster created by special interest groups fighting over their preferred way. It seems everyone hopes to win the fight by yelling at each other, including engineers like yourself. In the end the pollies throw their hands in despair and choosing the easiest option at the time. Everyone, and I do mean everyone including the engineers was wrong about that battery. It made money from the day it was installed. Turns out when a coal fired generator trips out and removes megawatts from the system in a single 50Hz cycle, to the computer controlling that battery that 20 milliseconds looks like an eternity. It can react in microseconds and dump compensating power into the system long enough for a peaker to fire up. And charge a small fortune for doing it. Apparently no one foresaw this, and so no serious grid scale batteries were added. Now everyone has seen they make money new battery installations are springing up like weeds all around the country. Again I'd love to say they are doing it for the climate or for grid stability but no, they are doing it to get on the gravy train. The way we are going about this transition is nothing if not consistent. Predicting what the end game looks like seems like fools errand to me. 100% renewable seems dubious. But they are at 70% now, so somewhere above 80% seems reasonable by 2030. Maybe they will start building pumped storage by 2050 - stranger things have happened. But using an SMR to fill that 20% gap - that's beyond strange. It's a shitty 20%. You are off most of the time. You have to go from 0 to full peak within an hour or so when the wind stops blowing and the sun ain't shining, and then from full peak back to 0 on a dime. Nuclear might be good at a lot of things, but the one thing it's absolutely hopeless at is load following. Japan having the worlds highest percentage of nuclear is also why Japan has the most pumped storage per unit generation. The economics aside, nuclear would require a lot of forward thinking and commitment. Clearly when it comes to power generation that hasn't been Australia's strong suite. I don't see it but maybe there is some place in the world that is different. (Who could possibly have thought buying gas from Russia was a good idea?) China would be a good candidate I guess. China does have a 30 year plan for building nuclear. But with the price plummeting on renewables it looks like they've now abandoned it in all but name as they are adding more renewables each year than the nuclear plan called for over it's entire lifetime.

hamilyon2 2y ago

What do you think about power plants apparently needing a lot of fresh, relatively cold, relatively clean water? A precious resource in future.

orson2077 2y ago
Modern powerplants are closed-loop and do not consume water, although they may dump heat into it. Water is not consumed (in any great quantity) or contaminated, except that which is recirculated inside.
- pfdietz 2y ago
  > although maybe they may dump heat into it. Dumping heat into water is how they consume water. They either evaporate the water directly, or they heat a river, causing it to evaporate more later.
cyberax 2y ago
Power plants can be designed to work in hot deserts. The largest nuclear power plant in the US (Palo Verde in AZ) gets cooled by evaporating treated wastewater. French nuclear power plants were just not designed for droughts.
schmidtleonard 2y ago
They need heat sinks, not fresh + cold + clean water. Even the heat sinks are only really necessary due to concentration of generating capacity rather than the amount of generating capacity. For example, photovoltaic usually has thermodynamic efficiency around 20% while steam plants (nuclear, fossil, geothermal, etc) are usually around 33%: solar panels will release considerably more heat into the environment per unit of energy generated, but since it's spread out nobody cares. Small Modular Reactors are a big step in the "spreading it out until it's easy to get rid of" direction. What does need (and not just "need" but actually "consume") fresh, cold, clean water (and dry air) is swamp cooling, which for some reason seems to do the rounds as as environmental silver bullet every few years. But that's a different rant.

DarkNova6 2y ago

Any design which you are favoring in general? From an outsider's perspective, SFR based designs look like hellish machines which are begging to burst up in flames. In contrast, I love the design of the properties LFR, except the material problems of high temperature lead. And in German circles the Dual Fluid reactor gets a lot of buzz as well.

orson2077 2y ago
I have a tremendous love for the fast breeder reactors (particularly Superphénix); that we had that technology so long ago astounds me. However, I recognise that they are technically challenging. From a modern, pragmatic point of view, I'm very partial to SMRs and AMRs (advanced modular reactors). They are also easier to implement on non-technical grounds (e.g. site permissions).
- pfdietz 2y ago
  France has basically abandoned their fast breeder effort. They defunded the next fast reactor.
cyberax 2y ago
Lead-fueled reactors are definitely an interesting area, and Russia is in the process of building a 300MWe demonstrator (BREST-300-OD). That being said, they have their own share of problems. Startup and maintenance is going to be a beast, slowly heating up the reactor to working temperatures is going to take almost a year. The other problem is that it's still not clear if corrosion problems can be solved satisfactorily. Reactor construction steel is stainless because it has a protective film of oxides on its surface, and lead gradually rubs it away (and then rubs away the construction steel, at a much faster pace). BREST-OD reactor designers spent a decade perfecting a system that is supposed to manage the amount of dissolved oxygen in lead, but it needs to be tested in the actual reactor conditions. With all its crazy temperature gradients and flows. Another interesting area is light-water cooled breeders. Such reactors _are_ possible, but just marginally. And they have the nasty positive void coefficient, but it looks like it should be possible to compensate for it.

matthewdgreen 2y ago

Sure. I have questions: 1. We need to be at net-zero by 2050 (and realistically based on the fast-moving temperature increases happening around us) maybe decades before that. SMRs haven’t been practically deployed yet, and don’t seem likely to be even minimally deployed until at least the 2030s if ever. How are we going to displace all forms of fossil fuel on the miniscule runway we’ll have left? 2. We have to deploy vast numbers of SMRs to the entire planet, including places with much worse security guarantees than first world nations. How do we propose to secure the huge amounts of fissile material and waste these reactors will use/produce. I’m not worried about fission bombs necessarily, but I am worried about pollution and dirty bombs. 3. “Without major improvements in storage” is doing a lot of heavy lifting. We’re seeing massive declines in storage prices and huge increases in production. If storage follows a curve similar to Solar PV, a huge fraction of the profitable applications for nuclear will be gone to cheap renewables and storage. Even if it doesn’t, renewables and today’s batteries are already driving fossil sources off the grid. How do we pay for a technology that will only have a few use-cases left after all the low hanging fruit has been eaten? PS The last question is not a troll. It’s very obvious that renewables are going to generate something like 80-90% of our power. I’m open to the possibility that nuclear could make up the remaining 10-20%. But the economics of that chunk will be messy, since SMRs will have to compete with dirt cheap electricity when the sun/wind are available (even if storage stays expensive.) I highly doubt that SMRs are going to outcompete Solar PV when the sun is shining. What do the economics of the mostly-renewables-and-storage-with-SMRs-as-backstop world look like?

ants_everywhere 2y ago
> I’m not worried about fission bombs necessarily, but I am worried about pollution and dirty bombs. Realistically, the nuclear industry will just decide in a few years that it's too hard to properly dispose of waste and they'll dump it unsafely somewhere that it harms humans. This will be easier to do in countries that already have large hazardous waste problems. In the US it will require a great deal of lobbying, but at least 1/3 of the population will support it if they think it makes the economy stronger and sticks it to their enemies.
h0l0cube 2y ago
> We’re seeing massive declines in storage prices and huge increases in production. Just in the last 9 months I’ve seen LFP prismatic cells from China drop 50% in price (and about 10% in pack volume). Add to that the variety of grid scale storage tech emerging out of R&D into the real world, by the time a nuclear reactor built today is coming online, it will be obsolete. More importantly, as renewables reach overcapacity, we need fast dispatch ‘peaker’ plants, not baseload, which makes gas a better transitional power source than nuclear, and batteries the end game

monkaiju 2y ago

Wild to me how someone can open with something as ridiculous as recognising "the non-negotiable need of humanity for ever-increasing amounts of energy" and then be taken seriously. Its not only negotiable, its a requirement that we do not maintain such consumptive expansion.

orson2077 2y ago
Hey, I'd love to live like the elves in Lord of the Rings, but it's not gonna happen. The first world offloads its manufacturing burden to the third world, then criticises them for polluting the atmosphere with fossil fuels. Also, the majority of the world do not enjoy the standard of living we do in the first world, and they will want to. Like it or not, energy demand will increase. It is non-negotiable unless you think there'll be some great die-off.

fragmede 2y ago

What would you do with a billion dollars?

fragmede 2y ago

Do you think we'll crack fusion?

olau 2y ago

The "strong developments" in the grid and storage you speak of is peanuts compared to trying to make nuclear cheap. Big, powerful companies with brilliant physicists and engineers have worked on it for decades. Lots of prototypes, lots of designs that didn't work out in practice. Today, we don't actually need new tech for distribution and storage, although new tech is being developed and helps. Chemical batteries for small-scale time-shifting and hydro plus gas plus biomass for larger scale. In 20 years, the youth of today will have to decide if they want to get rid of the last fossil gas - long term storage is on the order of 10% of total energy needed in the studies I've seen. It can be substituted with gas made from biomass. In fact, it's already happening in the country I'm in. Distribution is just building more connections. I think working on improving the cost of underground connections would help, but it's doable with today's tech, and it is happening. If anyone's interested, there are plenty of academic papers discussing this, and also a bunch of more accessible articles by this dude: https://cleantechnica.com/author/mikebarnard/ https://cleantechnica.com/author/mikebarnard/ For instance, here's a recent one https://cleantechnica.com/2024/04/11/the-short-list-of-climate-actions-that-will-work-india-edition-transcript/ https://cleantechnica.com/2024/04/11/the-short-list-of-clima... And yes, it's a complex topic. And yes, there will be some pain points along the way - energy is important to modern society, and it is a large transition that will years to unfold.

kkfx 2y ago

A personal take: the point for SMRs it's not costs nor potential mass-production and so on but a simple thing, they can be moved. We are in a changing world, we know, at least some of us know, we don't really know the future, but we know many will migrate and this means wars but also the inability to made fixed land infra, like roads, rails, electricity grid etc. Long story short: we need to have electricity "with us, moving" and p.v.+storage can works for a large slice of inhabited land but not enough, so we need something similarly easy to locally deploy. What do you think about that?

_ph_ 2y ago

I am sorry, but I think most of what you wrote is plain wrong. Yes, I agree, the total amount of energy used by humanity is going up. Especially as there are so many nations on earth which so far only use little energy. On the other side, it will somewhat go down from the current state for many leading industrial nations. However, the bigger change will be how we use energy. In the last century the whole grid was optimized for a mostly constant load because that is what the then-used technologies, nuclear and coal, could do best. There were even big incentives given to customers to have a mostly constant draw of power. But now this changes. We have energy sources which produce electricity very cheaply, but not at a constant rate. And the experience shows, if the end customer is charged by availability, the consumption patterns change to optimize costs. This will be a big factor in the rollout of renewables. As the article shows, cost-wise the SMRs can't compete. Renewables are getting ridiculously cheap. And even more: every one can set up renewables. You can go and buy yourself solar panels and put them onto your house or into your garden. The same advantage applies to industry-scale deployments. Same with wind. The only disadvantage with wind are the permits, as wind generators are quite big. But those are still trivial compared to nuclear, as the wind generators are not dangerous in any other sense. Because they are cheap, the electricity markes are going to be flooded by renewables. That is basic economy. So the question is, how can we complement renewables best, especially to cover the gaps in their production. It won't be one thing, but a combination of several options as they are non-exclusive. Local conditions will put a stronger emphasis on one technology vs. the other. But one thing is clear: it won't be nuclear. Because even in the most optimistic szenarios, nuclear is a bad counterpart, as it is not good at providing varying output. Even if that is technically possible (usually it is not), economically it doesn't make any sense at all. For a transition time, it will be gas, as gas plants are relatively cheap and fast. Gas is expensive, so running the plants infrequently makes sense. The gas can later be provided out of renewable production. Most likely, battery storage will kill that too. But if not, the gas power plants can still be used. And in all of that I haven't even touched the operational safety, nuclear waste, and of course giving nuclear technology to countries we don't trust.