Hacker News Reader

Comment by mcot2 | Hacker News Reader

Debatable if Nuclear is even required. Nuclear is fantastic for baseload but the economics are currently terrible mostly due to uncertainty in regulation and spiraling decommissioning and cleanup costs. We need much more investment in reactor technology and there is currently zero appetite for it in the United States.

There is however tremendous investments in battery technology. It has yet to be seen whether that could get cost effective enough combined with wind and solar to completely get rid of the need for Nuclear.

View on HN

Honestly it provides a huge amount of the base load in Ontario (~60%) and for most intents and purposes is clean. Debatable if we'd be able to get new reactors built today but I'm glad it exists

New reactors were considered for Ontario, but the plans were withdrawn because it didn’t make economic sense.

It would have been great if we had built enough nuclear plants in the 20th century, but at this point it’s pretty clear that most countries would better off going straight to cheaper renewables.

http://nuclearstreet.com/nuclear_power_industry_news/b/nucle...

> These are business decisions unique to Ontario and reflect the current realities of the market.

This doesn't seem to support your general argument.

Do you have any evidence that renewables are generally cheaper than nuclear when accounting for storage?

No, I’m not an analyst who crunches those numbers. But it’s worth highlighting that Ontario didn’t find it worth building more reactors, even with all of the necessary infrastructure in place and plenty of public support for the nuclear industry. And in the 10 years since then, renewables have only gotten cheaper.

How about asking what conditions would be needed in a different country to make new reactors much more feasible than in Ontario?

> Solar and wind became cheaper than competing new-build power plants years ago. What the latest report shows is that they have actually gotten so cheap that they are now competing with existing coal and nuclear power plants. In other words, new wind and solar farms can be cheaper than continuing to get power from existing coal and nuclear power plants. [1] [2]

> The storage report also shows a rapid drop in the costs of batteries, which leads to wind + storage or solar + storage getting increasingly competitive (and putting natural gas peaker plants out of business). [1] [2]

8 minute energy's blended solar and storage costs are roughly $40/MWh, just above existing nuclear generation per MWh costs (that doesn't fully internalize waste disposal costs and other ancillary costs of that power). “The vast majority of our development pipeline are solar-plus-storage projects, with more than 24GWh of energy storage under development,” 8minute Solar Energy’s Sean Kiernan said. [3] Portugal just signed a deal for a 700MW project at 1.3 cents/kwh, a historic low (mentioned to demonstrate the cost decline continues, and can be expected to continue into the future). [4]

If you're looking for storage that discharges for days, that is not necessary with robust transmission infrastructure (which needs to be built anyway due to aging infra in the US) and overbuilding of cheap (and rapidly cheaper) renewables [5].

[1] https://www.utilitydive.com/news/renewables-continue-to-get-...

[2] https://cleantechnica.com/2019/11/22/solar-costs-wind-costs-...

[3] https://www.energy-storage.news/news/developer-8minute-says-...

[4] https://www.pv-tech.org/news/52925

[5] https://news.ycombinator.com/item?id=23580707 (Hacker News: The US can reach 90% clean electricity by 2035 without increasing consumer bills)

I've said this before but there is no evidence that battery storage can scale to anything like the required storage. Tesla's battery in Australia stores roughly what the UK receives from French nuclear over production in a few minutes.

"no evidence that battery storage can scale"

Obviously laws of physics do not bar it from scaling, the question is around cost.

Are we short some raw material or why do you mean it couldn't work?

"As of 2016, global lithium-ion battery production capacity was 28 gigawatt-hours, with 16.4 GWh in China."

UK current electricity demand is 30GW. So the demand for one hour is similar to the entire lithium battery production for a whole year.

How many hours worth of storage would we need to balance the energy grid using only battery storage? Off grid homes usually aim for around 3-7 days worth. Obviously that's excessive for a whole country where weather conditions vary across the country but I'd think several house would be required to account for peak in demand at times of low production.

Realistically distributed storage and load side control I think are more likely to be a bigger part of making it work.

That's a fairly old statistic to pick, not sure if it was intentional or not, but the reality is lithium battery production has grown significantly since 2016 and shows no signs of stopping. The largest manufacturer in China, CATL is estimated to grow its capacity beyond 307GWh by 2028 [2].

"In 2019, LG Chem had the most lithium battery production capacity at over 50 GWh. LG Chem [is] increasing EV battery production capacity to as much as 110GWh by the end of 2020."

The design of a modern grid with battery storage would change as well - no reason to have singular giant storage plants. A mixture of solar, wind, and hydroelectric with distributed storage doesn't need to self-sustain 30GWh of storage, it needs to be able to smooth distribution and handle peaks. It'd be far more honest to look at seasonal average loads versus peaks, plus a buffer.

[1] https://energycentral.com/c/ec/world-battery-production

[2] https://www.statista.com/statistics/1103401/predicted-lithiu...

The UK has very little hydroelectric power and the undeveloped hydro resources are in remote areas of outstanding natural beauty (which is a shame because it's by far the best option for base load renewables).

The UKs current installed pumped hydro resources amount to some 28GWh. Clearly we need a multiple of that if that is currently required to balance a grid consisting of largely dispatchable power. Given that, I don't think my estimate of needing several hours worth storage is too far out.

That was the first statistic I found. Even if we say current production is 200GWh of lithium battery storage, that isn't going to go very far if the rest of the world also need to create similar storage infrastructure.

It would work just like running your house on a car battery, for about five minutes. My dad has 24 industrial lead acid 2V 450Ah cells which amount to 1m³ and 4.5kWh of usable storage since if you drain more than 20% you will kill the batteries too early. The 1140Wp PV array charges about 6kWh on a sunny summer day. We use about 5kWh daily in our apartment - that's no AC or electrical heating, just the basic appliances and a gas heater. He probably uses two to three times that amount, again no AC or electrical heating. Also lead acid batteries are rated for like 10y, PV retains 80% of its capacity for 25y, so you need to change 2.5 batteries during its lifetime. Batteries alone amount close to 50% of the TCO.

So good luck running the grid on batteries for more than 15 minutes. We'd probably need to cover Alaska with lithium batteries.

At least 40% base load is what is needed and one only gets that from nuclear and geothermal if we discard fossil fuels. Maybe solar thermal with molten salt storage too, dunno.

Lead Acid is pretty old tech. My car (Tesla Model 3) has a roughly 75kWh lithium-ion battery. Even with AC and/or a heat pump, that battery could run a rather large home for more than a day on it’s own. Their powerwall home battery is a roughly 17kWh battery that is stackable to build any size system you want.

Point of use energy generation and storage is a much bigger deal than I think people realize. I believe internally Tesla thinks this is many times it’s car business.

Nuclear is carbon clean yes, However it is by far dirtier than any other generation method, if we stop the coal plants today, in few hundred years CO2 levels will revert back and temperatures will normalize. The half life of the radioactive waste is in ten of thousands of years.

We are still struggling store industrial chemicals for mere half dozen years without it causing tragic accidents, while Canada might not be Lebanon today, there is no way to know how well off our future generations will be and whether they can safeguard the waste, both safely and securely from hostile actors without it being an big burden on them .

It will be costly problem at best or a major disaster at worst for all live forms. all this to have some power for 30-40 years.

If the world continues on a high-emissions trajectory, and people do not later pursue active remediation measures to remove CO2 from the atmosphere, it will take hundreds of thousands of years for natural silicate weathering to draw down atmospheric concentrations to preindustrial levels. More than half will be removed by faster processes, within 1000 years, but a significant minority of the excess will go away over the course of ~400,000 years:

https://www.nature.com/articles/climate.2008.122/figures/1

This is actually longer than the time it will take for the majority of fission products and actinides generated in nuclear fuel to decay. The waste CO2 problem is also distributed over the whole globe rather than easily confined to storage casks like spent nuclear fuel is.

Coal is bad, Co2 is bad, the idea is not the coal is better, we have more than two options, no point in jumping from frying pan into the fire.

Neptuimum-237 has half life of 2.144 Million years and is produced as nuclear fuel waste and one of most mobile nuclides at the Yucca repository. It will take 100's of millions of years for sizeable quantity to go through its decay chain and be safe.

There are safer better options than nuclear to coal. A 1000 year time line of natural restoration is not acceptable and needs to be addressed, but trading it for millions of years is not the right solution.

> "in few hundred years CO2 levels will revert back'

Where do you imagine billions of tons of CO2 will go? It does not have a half-life like nuclear waste, once you've taken that carbon out of the ground, it is here to stay for tens of millions of years. There will never be enough trees to sequester this carbon. We must expend trillions of dollars to reprocess it back into oil or rock-like formations.

Also in the entire history of nuclear industry there has been no credible case of 'hostile actors' coming close to stealing nuclear waste. That's because doing so doesn't make any sense - it is useless to nations, and if someone like Taliban tried to handle it, they'd probably die in the process.

Nuclear waste is not a liquid that might spill and run away if you aren't careful. When properly handled it is vitrified in solid glass, and it doesn't go anywhere. Sweden is placing it in underground storage in solid rock, in a region that's geologically stable for millions of years - it will never bother anyone again or cost a penny more.

>However it is by far dirtier than any other generation method,

Where do you think the fancy chemicals in the fiberglass turbines or the exotic metals in the solar panels comes from?

>We are still struggling store industrial chemicals for mere half dozen years without it causing tragic accident

Oh boy are you going to love the high tech chemical industry that underpins the manufacture of wind turbines and solar panels.

All these things are dirty if you don't clean up after them. Nuclear is pretty clean just by virtue of how little material you need to process per kilowatt. A kilo of uranium gets you a lot more energy than a kilo of turbine or a kilo of solar panel.

They're all better than fossil fuels because they don't crap byproducts into the atmosphere but you need to have so much less underlying industry per energy generated with nuclear that it's just inherently cleaner than anything else. This is assuming that you're cleaning up after whichever one you choose.

If you fully recycle the nuclear waste, the radioactivity drops to that of natural uranium ore in about 500 years: https://whatisnuclear.com/waste.html

Radioactive waste with half lives of tens of thousands of years is by definition not particularly radioactive. It is however reasonably well contained; roughly a truck load in a year. Where as CO2 is defused through the atmosphere in a way which makes it very difficult to recapture.

> Radioactive waste with half lives of tens of thousands of years is by definition not particularly radioactive

Sure plutonium-239 with half life of 24,000 is not particularly radioactive, neptunium-237 has half live of more than 2 million years, both are extremely dangerous to humans for multiples of that time.

Half live is not directly related to the level of damage gamma decay can do to you. Elements are perfectly capable of emitting hazardous amounts of radiation for thousands and millions of years time.

We cannot barely build a phone to last 3 years, The incentive to not create problems for future generations is simply not there. If it was, we would doing a lot more for pollution and plastics and the mass extinction event we are currently causing.

I don't believe our track record as a species warrants any trust on the ability to build containment to last thousands of years.

Plutonium-239 is an alpha emitter. Nuclear missiles on submarines actually have to spend more money to create highly pure Plutonium-239 specifically because it's long half life means it has a low rate of spontaneous emissions. This makes it safer for crew who have to live and work in close quarters with these missiles.

The double standard against nuclear always amuses me. If we truly want it to be similar to fossil fuels, we can just wait a couple hundred years for nuclear waste to cool down, take it out, grind it into a fine powder, and literally spew over the globe, and it will barely register on anything other than Geiger counters.

Meanwhile, fossil fuels generate so much waste per kWh that it's impossible to contain it, so we just spew it over the globe at day one, changing the Earth's climate and driving mass extinction (and killing a million per year), but it's OK, because if stop using fossil fuels today and wait a few hundred years then things will be back to normal. Maybe.

The double standard for nuclear worries me.

Of all the ideas on nuclear waste disposal including launching on rocket into the sun, this is the most dangerous[1].

Firstly If you are replacing Coal and other methods of energy generation with Uranium the amount of waste is not small anymore ( it already is not small in any sense).

Cooling down the temperature is not the only problem. Neptunium-237 has half life of 2Million years!, plutonium 239 of more than 24,000 years, these materials will be radioactive for a longer time than we as species have been on this planet. Grinding them fine and spreading them everywhere may very well kill all life on this planet for ten of millions of years, it may never come back.

Coal is bad, nobody is saying it is better than nuclear, but replacing coal with nuclear is opening a whole new can of worms, we are not equipped to handle, there are plenty of other safer alternatives .

[1] In in interest of a civil discussion, I will leave it at that.

Appreciate the civil discussion, so let me just say that I suspect you never typed in the numbers and tried to quantify the danger of nuclear waste. Let's try a quick and rough estimate.

The energy density of uranium is roughly 80,620,000 MJ/kg [1]. Existing nuclear power plants have thermal efficiency between 30-40%, roughly speaking [2], so let's be pessimistic and choose 30%. That means actual usable energy density of uranium is 24,186,000 MJ/kg, or 6,718,333 kWh/kg.

The world produced 27,644,800 GWh = 27,644,800,000,000 kWh of electricy on 2019 [3]. If we used nuclear to generate all of this, we'd need 4,114,830 kg of uranium, or 4,115 ton per year. Let's round up to 5,000 ton.

Spent nuclear fuel contains about 0.8% plutonium 239 [4], so our 5,000 ton of uranium will generate about 40 ton of plutonium 239, with half-life 24,000 years and radioactivity of 2.3 GBq/g [5]. So our hypothetical nuclear waste will have radioactivity of 2.3 GBq/g * 40 t = 92 PBq. Which is a lot.

...or is it?

Potassium, naturally occurring and essential for life, is naturally radioactive (31 Bq/g), because it contains 0.012% K40 [6]. Seawater is about 0.04% potassium [7], which contributes to a natural radioactivity of 12.4 Bq/kg, or 12.4 kBq/t from potassium alone.

Actually, let's forget the earth. Let's spread the nuclear waste evenly on the Mediterranean - volume 3,750,000 km^3 [8] - who needs Italy anyway?

It contains 3,750,000,000,000,000 ton of seawater, or 46,500,000,000,000,000 kBq of natural radioactivity, which is 46,500 PBq. So our hypothetical dump of plutonium 239 will increase the Mediterranean sea's natural radiation level by about 92/46500 = 0.2% per year [9].

A far cry from extinction of all known life, I have to say.

* By the way, you may want to re-consider before saying stuff like "Pu239 has half-life of 24,000 years - it's not safe!" Coal, for example, contains tons of mercury, which is not radiactive (i.e., it has a half-life of infinity). After 24,000 years, there will be 50% of Pu239, but 100% of mercury left from coal ashes. Which is worse?

[1] https://en.wikipedia.org/wiki/Energy_density

[2] https://www.world-nuclear.org/information-library/current-an...

[3] https://en.wikipedia.org/wiki/List_of_countries_by_electrici...

[4] https://en.wikipedia.org/wiki/Plutonium-239#In_nuclear_power...

[5] https://www.radioactivity.eu.com/site/pages/Plutonium_Proper...

[6] https://en.wikipedia.org/wiki/Potassium

[7] https://en.wikipedia.org/wiki/Seawater

[8] https://en.wikipedia.org/wiki/Mediterranean_Sea

[9] Of course the actual effect will be higher because there will be other isotopes, shorter-lived and more radioactive. That's why I proposed keeping the waste in a fridge for 200 years first. I'm not a monster.

It kills the fewest number of people pre joule and the negative externalities are centralized, i.e., the people that made the mess need to pay to keep it stored safely. I fail to see how an underground bunker of radiation is dirtier than people literally breathing in toxic fumes. Coal even releases a ton of radiation per joule, so even the framing of radioactive vs non-radioactive is a subject of discussion.

Also, I highly doubt that if humanity makes it even a thousand more years that we'll have trouble finding a productive use for depleted uranium or other spent fissile fuels. They have many interesting properties.

The people and the governments who took those decisions will be long gone in few hundred years, the waste will still be here , leaking into the soil and water table.

I was not arguing for coal in any way I am merely pointing saying Nuclear is not as clean as pro nuclear people argue.

Yes the NORM released by Coal and other human activities is substantial, However it is not remotely comparable to Fukishma, Chernobyl and bikini atoll . I don't trust any of the world governments to build a safe reactor and I don't share your optimism on that we will find ways to handle the waste. Instead of trusting luck that we will find clean ways to handle, we can start building plants when we do find them, .

It is recency bias to think we will be able to keep technologically innovating at the same pace we have in the last few decades or even few hundred years, most of our history innovation has happened rapidly only under certain conditions, it may be that those conditions will continue to favour, or maybe physical limits such as moore's law slowing down, resource limits stuck on single planet economically or socio political structural problems can slow down innovation substantially, Any of them could be a great filter in the Fermi Paradox sense.

> the people that made the mess need to pay to keep it stored safely.

We'd certainly like for those people to keep the mess stored safely, but there are incentives for them to not do so, even assuming they had the ability to.

Nuclear waste is so so so small, it's okay for it to last a long time.

And you'd better not ignore how much radiation is released by coal.

> [...] and for most intents and purposes is clean.

It is not:

https://www.sortirdunucleaire.org/Nuclear-power-a-false-solu...

Even the best batteries in the world have absolutely ridiculously small energy densities compared to Uranium (about a factor of 10^6).

Energy density almost doesn't matter for grid scale things. It's not like we don't have space where we could put batteries. Cost is really the only important measure and nuclear is very expensive, while batteries keep getting cheaper.

Energy density matters greatly because cost is a function of the material you have to handle and the amount of facilities you need to build to generate the power in the first place.

In that case energy density would be reflected in the cost. That is why cost is often used to compare energy technologies, and not cherry picking a random metric which favors one particular technology.

The cost of nuclear reflects the fact that toxic waste you can't see, smell or taste is scarier in the minds of the public so we insist upon far more government involvement and rigor in the construction of nuclear plants and insist they pay their decommissioning costs up front. Dare I say nuclear's costs are internalized whereas other modern manufacturing's costs are externalized.

They don't need a whole nuclear power station built around them though.

They also have orders of magnitude (10^3 - 10^4) more charge/discharge cycles than uranium.

This is a joke, right?

Its an observation that comparing vastly different technologies using just a single metric produces mostly misleading asinine statements that aren't sufficient to influence any meaningful policy or engineering decisions without additional supporting explanation and justification.

Right, so it is a joke :)

Current nuclear is so expensive and has so many drawbacks that new nuclear plants can't compete with solar / wind and storage combination.

However, that's not an innate property of nuclear, and there are relatively unproven designs that may be sufficiently cost effective and safe to be a desirable part of the mix.

I think the free market will solve this problem better than is speculating about it on HN. If its cost effective, it will happen, if it's not, it won't.

Seems like it's a function of the free market and politics—there is a spectrum of regulations we can have that ranges between "enough to keep nuclear energy safe" to "enough to make even the best nuclear technology non-viable", and where we fall along that line is primarily a political question.

Yes, that is true, and the regulatory environment is a big part of why current nuclear tech is non-viable.

> new nuclear plants can't compete with solar / wind and storage combination

There dose not exist to my knowledge (would be happy to be proven wrong) any storage solution that currently operated commercial from buying cheap renewable energy at peak production and regain investment and operational costs by selling it expensive during lows. Not a single one in the world.

At best any comparison between storage and nuclear plants is between two currently non-viable commercial alternatives, both requiring massive government funding to be built. The only commercial viable power source right now is renewable that can compete against fossil fuels when the weather make it viable, and fossil fuels as base load and as a dynamic source that can spin up based on weather conditions and demands.

You quoted a part of my comment and then ignored it in yours. Fossil fuels don't even enter into my comment, I'm merely pointing out the the multiple billions of dollars and multiple years required to create a new nuclear power plant make it non-viable when compared to the current cost of renewable power and grid scale energy storage. These are publicly available numbers, you are free to look it up and do your own calculations if you doubt me.

Also, you're simply unaware of commercially successful energy storage. The most high profile example being perhaps the battery pack Tesla installed in Australia for $90.6M AUD, and generated $13.1M AUD in revenue in just six months[1]. Clearly a very good ROI. Pumped Hydro is used all over the world and is even more cost effective.

[1] https://www.theverge.com/2019/6/25/18715585/tesla-australia-...

Im trying to find a source that the battery pack Tesla installed is used to buy cheap renewable when the price is low and sell when it is high, but the linked article does not state this and the article it link as source for the 13.1M AUD is behind a paywall.

Pumped Hydro like Bath County Pumped Storage Station does exist but do not seem to buy cheap renewable when the price is low and resell it when it is high, but rather operate as a balancer for PJM Interconnection which get almost all its power from fossil fuels.

So do you have a link for a profitable battery that buys energy from renewables when there is an overproduction from renewable sources and then later sold when the prices has risen? That would be a solar / wind and storage combination. Using energy from fossil fuels to pump water is just not the same, even if doing so creates a profit for the company.

Every time I have read about a new storage facility they usually start with first describing capacity, then go on to say that this can enable a future use with renewables, but that right now it operate to balance the grid of existing (mostly fossil fuel based) power plants. I would very much like to be proven wrong and that there exist a viable commercial operation where we can see X amount of green energy produced by wind and solar going into the storage, and X (minus a conversion cost) of that energy going out, with one table showing how much the green energy costed when going in and an other table showing how much they got in profit when it went out.

Why would you ever build a company around arbitraging power and all the risk involved, when instead you could just buy panels or wind turbines and skip directly to selling power?

I don't know, ask Elon:

"Tesla applies for UK power generation licence in [virtual power plant] play. ... The move suggests Tesla may be planning to build large-scale battery storage projects in the UK"[0]

[0] https://theenergyst.com/tesla-applies-for-uk-power-generatio...

Arguably nuclear is only expensive because we've lost the skills to build it at scale and quickly, I the 70's it did not take 15 years to build a power stations, and neither does it in China today - they are building reactors at unprecedented scale, cost effectively and on schedule.

In 2009 their intention was to "raise the percentage of China's electricity produced by nuclear power to 6% by 2020"

In 2019 nuclear produced 4.9% of the gridpower.

https://en.wikipedia.org/wiki/Nuclear_power_in_China

This is hardly an "unprecedented scale", which AFAIK lies elsewhere: https://unfccc.int/news/china-and-india-lead-global-renewabl...

Yes, much like the US has forgotten how to build subways, tunnels, bridges, and infrastructure cost effectively in general. I don't know how that happened, and I think it's a pretty sad outcome.

Maybe in a different regulatory and economic environment, nuclear could be competitive. I don't know. Even coal is becoming uncompetitive in the US compared with renewables, and it's a lot cheaper and lower risk than nuclear.

Reply on news.ycombinator.com