Hacker News new | past | comments | ask | show | jobs | submit login

The current price of nuclear power in the US and Europe is almost entirely due to overregulation combined with decades of dwindling expertise, with a sprinkle of corruption on top. If we remove this regulatory sabotage, nuclear energy can be produced at about ~$30-40/MWh [1] (At 3% discount rate, and in the case of the US/Europe with the assumption that excessive regulation is scaled back a notch.). If we build enough, we could bring that below $30/MWh from the efficiency of scale.

Green hydrogen tends to cost ~$100/MWh, just for the fuel. Building and operating the plants themselves comes on top of that.

Even if the fuel comes down to $40/MWh by 2050, as predicted in this article[2], it will still be competitive with properly organized nuclear.

[1] https://world-nuclear.org/information-library/economic-aspec...

[2] https://www.energytech.com/renewables/article/21238510/ridin...




> The current price of nuclear power in the US and Europe is almost entirely due to overregulation

And here I thought it was actually due to the manufacturers living in Fuckupstan, not being able to build their products at costs they had promised. The recent French experience with the EPR is astoundingly bad.

Of course that cost could be lower if any screwup could be excused by greasing the appropriate regulatory palms. The regulation would be "over" in the sense that it adds to cost, but is it "over" in the sense of not being necessary? You do not have the contrafactual evidence of what nuclear safety would be like without that regulation.

The first link you gave there presented assertions that French reactors were cheap back when they were building lots of them. But those cost figures are opaque and unauditable and cannot be used as evidence that the reactors actually cost what they claimed. Those making those claims had every motivation to lowball them.

It's also improper to assume nuclear and renewables are assigned the same discount rate. Nuclear presents larger risks to investors (from technological obsolescence and from failure to complete the power plants at all), which should properly be accounted for by imposing higher interest rates. Assuming renewables and nuclear get the same interest rate is an implicit subsidy.

Ultimately, the pronuclear antiregulatory position boils down to "why don't you just suck it up and let us subject you to more radiation, which really isn't bad? You're being so unfair!" One sees this in the whining about the LNT hypothesis for radiation effects.


> Ultimately, the pronuclear antiregulatory position boils down to "why don't you just suck it up and let us subject you to more radiation, which really isn't bad?

Well, actually, yes. Coal plants cause 100 times more radiation than nuclear plants per unit of energy produced:

https://www.scientificamerican.com/article/coal-ash-is-more-...

And typical background radiation is 10x more than even that.

And this is for plants built in the 1970s. Tightening requirements beyond that while still allowing coal power to continue their emissions, was completely absurd.

> You do not have the contrafactual evidence of what nuclear safety would be like without that regulation.

To the contrary. Most nuclear power in existence is still produced by plants that was built BEFORE many of the excessive regulations were put into place. That's part of the reason they were so much cheaper to build. (I'm not making an argument for building Chernobyl type plants, but rather plants with safetly levels corresponding to those built in the 70s in Western countries.)

> LNT hypothesis for radiation effects

The LNT hypothesis has little to no evidence supporting it over a null hypothesis that radiation below a threshold of about 80 mSv/y. Which is based on statistics for huge groups of people. And the typical radiation received per year from living next to a nuclear plant is around 0.001 mSv/y, 80000 less than the amount where we have any data to indicate that the exposure is harmful.

If you have 1 xray taken in a hospital, that's as much radiation exposure as living next to a nuclear plant for 100 years. If you have a CT scan, that's like 1000 to 10000 years. Or, if you fly from LA to NY, you get 0.035 mSv of radiation, the same as 35 years. Do a round trip, and it corresponds to a lifetime.

Whether or not you believe in the unfalsifiable hypothesis called LNT, the risk from nuclear power is incredibly tiny compared to ANYTHING we do.

Meanwhile, the potential benefits of cheap nuclear power would be massive, both in terms of local air polution (when it becomes cheap enough to replace coal and natural gas), economical benefits or global warming mitigation.

And it has a proven track record. France was able to almost completely cut fossil fuels for their electricity production over a few years, with a moderate investment into nuclear. Meanwhile, Germany has spent 100s of billions of € on "renewable" energy, but has almost as high percentage of their electricity produced coming from fossil fuels now as when they started. (They did have some reduction in CO2 emissions due to moving from coal to natural gas, at the cost of becoming highly dependent on Russia)

> You're being so unfair! ... whining about the LNT hypothesis....

This kind of emotional response makes it seem that something else is at stake.

Of course, if we were able to provide clean air and stop global warming using nuclear energy, some people would lose their jobs, both people selling other forms of energy (fossil, wind, solar) or people working for "environmentalist" organizations or political parties.


> If you have 1 xray taken in a hospital, that's as much radiation exposure as living next to a nuclear plant for 100 years.

Now do living near Serpent river, or Church Hill, or Kadapa, or Ranger or Mailuu-Suu or the cumulative worldwide effect if we got 15% of our power from a reprpcessing facility like La Hague. Make sure to include the effects of heavy metal poisoning, not just radiation. Also would you like to replace your drinking water with aquifer water from Inkai?

> the potential benefits of cheap nuclear power would be massive

Here's a riddle. What fraction of world energy can 40,000t of fissile material provide and for how long? Where do you propose to get more?


> Make sure to include the effects of heavy metal poisoning, not just radiation.

I'm a bit lazy. Maybe you have some numbers available for this? My guess is that deaths caused by this (associated with nuclear power), if at all measurable, would be orders of magnitude lower than deaths from the extraction and pollution associated with fossil fuels (Per GWh).

> Here's a riddle. What fraction of world energy can 40,000t of fissile material provide and for how long? Where do you propose to get more?

There is around 40 trillion tons of uranium in Earth's crust. How much of that we can utilize, depends on how much we're willing to pay for the extraction. The current fuel price (after processing) is about $5/MWh, and reserve estimates are based on that price. Should fuel prices go up a bit, more mines and excavations will be profitable.


> My guess is that deaths caused by this (associated with nuclear power), if at all measurable, would be orders of magnitude lower than deaths from the extraction and pollution associated with fossil fuels (Per GWh).

The competition isn't fossil fuels. Everyone wants to get rid of them.

The competition is renewables.


> The competition is renewables.

When those regulations were introduced, the only renewables were hydro. And nobody was worried about the tiny amounts of radiations coming from coal plants, it was the soot that killed people.

Still, the "green" movement in Europe have been fighting nuclear power since at least the 80s, usually with more fervor than they've been fighting fossil fuels. The nuclear scare must have been easy to sell (and so an easy source of contributions), especially in the years after Chernobyl. With catastrophic effects both for the local environment and the climate.

Utopianists may indeed see nuclear as a threat to their dream of a perfect world. To me, nuclear is simply one of several energy sources with very low impact to the environment and climate, one that we _could_ have elected to produce at a low price. And still can.


Well good thing on river hydro is a separate category from renewables and noone is suggesting going back to it.

All of the examples of cheap nuclear power are ridden with corruption scandals and incredibly unreliable.

If you decide the CCP are suddenly trustworthy and ignore that finance and insurance have costs then the very limited fraction of nuclear power that can be produced might be both, but that doesn't make mining uranium any less horrific.

Cry bullying about the mean greens that have never been in power and only rarely held minority coalition positions just makes you look pathetic.


You're talking about scifi. I'm talking about the next 8 years.

You're also completely failing to distinguish between fertile and fissile.

So realistically, at the same scale of ore extraction as coal mining. How much power is available?


> So realistically, at the same scale of ore extraction as coal mining. How much power is available?

This article estimates that regular supplies is enough for 230 years at the current rate of consumption:

https://www.scientificamerican.com/article/how-long-will-glo...

If we triple the the number of plants, that would be approximately a 75 year lifetime for existing plants.

Beyond that, the main directions to take would be to use breeder reactors, which would be enough for 30000 years at today consumption, using regular fuels or to extract uranium from seawater, which would provide enough uranium for 60000 years at present rates (and much more if combined with breeder reactors).

In total, there is enough uranium to last thousands of years, even consumption goes up 10x or more.

Obviously, costs will gradually go up, or at least the extraction will require more advanced technology. Even just 75 years is a long time, and a lot can change by year 2100. Thorium or fusion power could be solved by then, or we could have space based solar covering our needs.


I said realistically. And this decade. Not in some weird scifi scenario where we have enrichment facilities that are 10x more effective and efficient and you have the reprocessed waste before you start.

Tripling generation by adding 600GW is nothing.

8 years.

2TWe net installed by the end of it and producing around 300GWe net of new capacity per year.

Renewables are on track (and 2TW is extremely pessimistic). What's your plan? How much ore? Where is the scale?

Ignore the cost. You gotta demonstrate it's possible before you can gaslight about costs.


> 8 years.

8 years is barely enough to start changing course in how things like energy production is organized.

Is there a specific reason you insist on this kind of velocity? Global warming is going to gradually increase as a problem over the next 200 years, if we continue our current course, it's not like the world is ending in 2030. In fact, on our current trajectory, the truly hellish outcomes are not expected until around 2150-2250 (based in IPCC reports).

But precisely because it takes so long to change course, we need to start turning the ship now.

> Tripling generation by adding 600GW is nothing.

Watts is not a unit of energy, it's a unit of power. Peak capacity is not very interesting. What matters is actual production as well as the cost of producing the power when there is demand (including the cost of storage, if needed).

Nuclear produced 2.8TW last year, which is roughly identical to Solar+Wind. That's about 10% each. If we add hydro to this mix (currently 15%), we have a total of 35%. The depressing part is that this has been relatively constant since 1985, meaning we haven't made any progress over the last 37 years.

https://ourworldindata.org/electricity-mix

However, if we restart investments in nuclear, while continue our renewable investments, we may be able to triple both over the next 15-20 years. If hydro remains constant, we may produce enough energy to cover 75% of 2021 consumption by 2040 (which would perhaps be 50% of 2040 consumption). That should be enough to replace most fossil fuels for electricity in the EU+US, at least.

To reach such a level is highly non-trivial, both for nuclear and wind/solar. For nuclear, it means a u-turn is needed on several fronts, and for wind/solar, there are economic and geographic limitations (some areas are getting saturated).

For this plan and time horizon, there are plenty of uranium deposits that can be mined, perhaps most interestingly for western countries, a lot of this is in Australia and Canada.

If we restart construction of nuclear plants, we will also drive incentives to go looking for more sources elsewhere.

As for your "Peak Uranium" hypothesis, I suggest looking at the history for "Peak Oil". Oil was predicted to reach its peak in year 2000, but the reality is that production is still increasing. I would be very surprised if the same is not happening for uranium.

And as far as I can tell, we DO need it (or nuclear in some other form), if we're supposed ween ourselves off fossil fuels this century. Wind and solar may be competitive in some locations up to some production volumes, but they seem to have very diminishing returns above some level, due to storage costs and available land areas.

Maybe we can, some day, have solar panels carpeting the Sahara or even in Outer Space, but that's definitely scifi.


Net watts are a measure of average capacity and are not peak watts. TWh per year is a unit of net power. But you know this and you know the figures I was quoting were net because you know the output of the world's nuclear fleet and you know renewables are slightly higher.

Just as you said: renewables exceeded the nuclear fleet last year, growing by 50GW net. Production capacity is online for another 100GW net this coming year. China alone has an achievable plan for half a terawatt of new net production, and renewable targets have been consistently exceeded.

> Is there a specific reason you insist on this kind of velocity? Global warming is going to gradually increase as a problem over the next 200 years, if we continue our current course, it's not like the world is ending in 2030.

This kind of velocity is the pace the renewable industry is operating at, with a clear roadmap to meet the target, and the pace it is necessary to move at to avoid the worst outcomes.

If the nuclear industry can't scale to meet it, that's fine. We'll use the technologies that can.

> we may be able to triple both over the next 15-20 years.

So you're saying if we invest heavily in nuclear it may be able to contribute 5% of primary energy in 15 years? Wind is on track to triple in under half of that, solar in around a quarter. Both are on track to provide a meaningful portion of primary energy in 15-20 years.

> Nuclear produced 2.8TW

TWh. Which is around 320GW.

> As for your "Peak Uranium" hypothesis, I suggest looking at the history for "Peak Oil". Oil was predicted to reach its peak in year 2000,

And drilling has gotten more destructive and energy intensive ever since. Oil and gas platforms are resorting to using nuclear, solar, and wind to keep extracting because oil is not a sufficient energy source to extract oil. The predictions about the resources were accurate. The predictions that we'd lean into the insanity of continuing extracting ultra deep oil or tar sands when it's barely energy positive are where it went wrong.

> And as far as I can tell, we DO need it (or nuclear in some other form), if we're supposed ween ourselves off fossil fuels this century. Wind and solar may be competitive in some locations up to some production volumes, but they seem to have very diminishing returns above some level, due to storage costs and available land areas.

Very nice weasel words. I've never asserted that nuclear can't contribute, only that it cannot match the scale of renewables and suggesting we stop renewable investment to focus on it because only nuclear can scale is a blatant lie that serves only delay decarbonization. You've just reasserted that this is true. Thank you for agreeing.

> Maybe we can, some day, have solar panels carpeting the Sahara or even in Outer Space, but that's definitely scifi.

Revealing further that you can't comprehend how renewables scale. As a demonstration of how terrible a representation of scale this is:

World primary energy is about 17TW or 2kW per person. In the regions that 93% of people live, this takes under 50m^2 per person. There are a few cities like Milan with more people than sunlight, but there is enough space in Tokyo to provide this much net energy for every resident and still have plenty left over for outdoor spaces. The denser regions can import energy heavy goods, and still have enough space for electricity if they really didn't want to put a few shades up on some livestock farms.

Simply covering the space rendered uninhabitable by Inkai Uranium mine would provide more energy than the mine does.


> This kind of velocity is the pace the renewable industry is operating at, with a clear roadmap to meet the target, and the pace it is necessary to move at to avoid the worst outcomes.

I disagree. The "worst outcomes" are 200 years into the future, and a ramp up speed of 10 years doesn't matter much for that.

> So you're saying if we invest heavily in nuclear it may be able to contribute 5% of primary energy in 15 years?

There are different ways to calculate "primary energy". Adjusted for inefficiencies, nuclear is 4.3%. In other words, tripling that means we can shut down at least ~9% of PE worth of fossil fuels plants.

Renewables get a similar boost from this approach, of course, at least long as we don't have to store it.

https://ourworldindata.org/energy-substitution-method

> So you're saying if we invest heavily in nuclear it may be able to contribute 5% of primary energy in 15 years?

No, I'm saying we reduce the unneccesary costs, and let it pay for itself. By comparison, Germany has to impose a 25% "green energy" tax on electricity (including nuclear) to stimulate renewables.

> Oil and gas platforms are resorting to using nuclear, solar, and wind to keep extracting because oil is not a sufficient energy source to extract oil.

Oil is more valueable as a transportation fuel than as fuel for electricity production. And extraction uses electricity. This is about market price, not EROI. (Also, for instance in Norway, it's about CO2 quotas. Norwegian oil platforms are moving the land based electricity instead of the natural gas they extract alongside the oil for their electricity needs.)

EROI for nuclear is still around the highest there is, around 100x. There is massive headroom before EROI for nuclear goes down to unviable levels. (3x)

> .... weasel words ... because only nuclear can scale is a blatant lie ...

I didn't say only nuclear can scale. I do claim that nuclear is a better source of energy when it's dark and there's no wind.

Also, ad hominem attacks doesn't help your case.

> Revealing further that you can't comprehend how renewables scale.

More ad hominem. Do you want to start a flame war?

> 2kW per person. this takes under 50m^2 per person.

Maybe you should re-read your sources. Pretty sure you will find that 2kW is around the average output of 50m^2 during the peak of the day. This illustrates a risk of measuring energy in watts. Most such calculations use 4-6 as estimates for number of "hours" worth at such an output, meaning the area needed goes up by a factor of 4-6. So let's say 250m^2.

Now, on top of this, the energy tends to be needed either in a different location or at a different time. Batteres with a 70% efficiency increase this to 350m^2 while storing it as H2 at 25% full-cycle-efficiency increases it to 1000m^2. Multiply by the number of people on Earth, and you get a square of 2800 km on each side (8 million km^2). Which is close to the size of the Sahara.

That's all if you're planning to use the energy in the same location, and not transporting it anywhere.

To be fair, this would be electrical energy, which has higher value than the average primary energy. So only half the size of the Sahara (maybe 1/4 if it's located in the ACTUAL Sahara, since that place is rather sunny.)

On the other hand, world energy consumption is going up every year.

Btw, unless you put away those ad hominem attacks, I'm not going to reply further.


> I disagree. The "worst outcomes" are 200 years into the future, and a ramp up speed of 10 years doesn't matter much for that.

So the renewable targets (which are being met) need to slow down and wait for nuclear energy which is somehow necessary to meet those decarbonization targets which ... would then result in not meeting those targets but that's fine because they're too aggressive? Sounds almost like the goal is to delay partial decarbonization by claiming there is a better solution later.

> There are different ways to calculate "primary energy". Adjusted for inefficiencies, nuclear is 4.3%. In other words, tripling that means we can shut down at least ~9% of PE worth of fossil fuels plants.

> Renewables get a similar boost from this approach, of course, at least long as we don't have to store it.

So if you ignore all the non-low-grade heat and inefficiencies entailed in turning electricity and low grade heat into chemical feed stock and the countries in energy poverty you can manipulate a number? Well done. Nice frozen world fallacy. 10% is still a tiny part of the problem.

Now after moving the goal posts 2/3rds of the way across the field, show some evidence that they can be met by demonstrating a potential contribution to a meaningful chunk of the problem. How do you get to 2TW of nuclear production in the same timelines as the renewable energy targets where does the Uranium come from?

> More ad hominem. Do you want to start a flame war?

Demonstrating ignorance or willful misrepresentation consistently on every single point that can be checked is more than sufficient grounds for requiring positive evidence for the claims for which your strongest argument is: 'you can't prove categorically that it's impossible for a solution to very obvious issues to appear later'.

> Maybe you should re-read your sources. Pretty sure you will find that 2kW is around the average output of 50m^2 during the peak of the day. This illustrates a risk of measuring energy in watts. Most such calculations use 4-6 as estimates for number of "hours" worth at such an output, meaning the area needed goes up by a factor of 4-6. So let's say 250m^2.

Nameplate watts aren't net watts. Everyone knows this. You know this, you just stated so. So double counting capacity factor can only be an intentional lie. 2kW peak would be a sixth of that with state of the art mass production panels -- on the order of 8.5-10m^2 or as little as 7.5 for bifacial panels with <100% coverage. Some utility systems have 50% coverage ratio, others have 98%, the 50% ones are usually optimized for more than the fixed tilt solar resource.

If you were covering an equivalent in urban land of a certain area in the form of walls, roofs, footpath shades etc. then by definition the area you are shading is the area you are collecting light from, so by shading a third of tokyo you can still make net exports from tokyo for a substantial portion of the residents' industrial production. The land use is both a non issue and smaller than the land use from Uranium mining.

> Now, on top of this, the energy tends to be needed either in a different location or at a different time. Batteres with a 70% efficiency increase this to 350m^2 while storing it as H2 at 25% full-cycle-efficiency increases it to 1000m^2.

Very few people live anywhere with less than 3.5kWh/day and the overwhelming majority of those who don't have existing nuclear and already developed hydro and wind resource. So around 40W/m^2 is accurate when sourcing mostly electricity and some low grade heat (this is very shocking, I know, but things get hot when left in the sun and you don't need to use an element and a PV panel to heat water or sand).

Even using exclusively winter sunlight from regions within AC transmission distance of >93% of the population would only double this.

30% battery losses are fairly old technogy or a system like PHES, direct thermal storage exists, you don't need all energy to go through seasonal storage as hydrogen and for every joule to be created in seattle during winter. You especially don't need hydrogen to be burnt or put into a fuel cell to create hydrogen for chemical feed stock or high grade heat. PEM electrolysers are much more efficient than alkaline and improving monthly. Hydrogen doesn't need to go through a rankine cycle steam engine to be used for electricity. Finally solar resource in a good area is closer to 80W/m^2 average than the 40 I used above.

If we needed every single joule to be from sunlight rather than as a salient example of how ridiculous the land use argument is then high energy intensity goods can just be created in sunny areas using PV and CSP (which is dispatchable) and shipped.

Care to try again but without the bit where every single number in your calculation is an intentional misrepresentation of current established technology (let alone emerging mass production technology)?

> Btw, unless you put away those ad hominem attacks, I'm not going to reply further.

Need an out to claim you're leaving because everyone is mean rather than because all of your bs has been called and you're out of new angles, huh?



Failing to understand the distinction between power and energy doesn't make you seem very credible. And neither does saying failing to understand what a breeder is.

You need fissile material to start a reactor of any kind.

Working breeders with a real closed fuel cycle don't exist but if we pretend they do it's about 5 tonnes per GW. You can't start breeding until they're built and the breeding ratio of proposed designs takes on the order of a decade to fuel another reactor.

How many billions of tonnes of ore do you need to extract the uranium from per year to meet net zero installed power roadmaps? How do you get to 2TW by 2030 to come close to the scale of the renewable roadmap?


They don't do it because fuel is so cheap, lol.


How much fissile material do you need to meet net zero by 2050 with all nuclear?




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: