Arguments
Is Nuclear Energy the Answer?
Posted on 13 June 2019 by scaddenp
Abbott 2011 and Abbott 2012 doesn’t think so but perhaps there are better analyses? For discussions of economics, levelized cost estimates of various electricity technologies can be found here and here.
Nuclear energy is quite commonly proposed as the solution to reducing GHG emissions. As soon as this gets raised on an article's comment thread, there has been a bad tendency for on-topic discussion to be completely derailed by proponents for and against.
We have repeatedly asked for nuclear proponents to provide an article for this site which puts the case based on published science but so far we haven't had a taker. The proposal would need to be reviewed by Sks volunteers. In lieu of such an article, this topic has been created where such discussions can take place.
However, in the absence of a proper article summarizing the science, stricter than normal moderation will be applied to ensure that all assertions made for or against are backed by references to published studies, preferably in peer-reviewed journals.
Update - October 2020
This post has been up for a little over a year now, and has received over 200 comments. Now seems like a good time to add some clarification.
First of all, the challenge to "nuclear proponents" to provide an article requires that the article "summarize the science". It is not the desire of Skeptical Science to provide a one-sided, pro-nuclear assertion. The expectation is that an article would provide a balanced review of all aspects of nuclear energy as a practical, affordable, realistic source of low-carbon energy.
If you think of yourself as a "nuclear advocate", then writing a balanced article will be difficult for you. This is not a place for "lawyers' science", where the role is to pick a side and pretend there is no other reasonable argument. This is not about winning an argument - it is about coming to a common understanding based on all the available evidence.
If you think that criticism of your position represents an "anti-nuclear bias", then writing a balanced article will be difficult for you.
If you think that you are the only one that truly understands nuclear energy, then you are probably wrong.
Review of any submitted article will not be at the level of a review of a professional journal article, but anyone submitting an article needs to be prepared to have their positions examined in detail for weaknesses, missing information, lack of support in the peer-reviewed literature, etc. If you find it tough to accept criticism in the comments thread, then you will not find review any gentler.
The problem with solar plus storage in the LCOE from Lazard is that the basis is 4 hours storage x 50 MW = 200 MWh for a 100 MW solar plant. It is not steady base load, but an "energy storage system designed to be paired with large solar PV facilities to better align timing of PV generation with system demand, reduce curtailment and provide grid support." Technologies assessed were Lithium Iron Phosphate (LFP) and Lithium Nickel Manganese Cobalt Oxide (NMC) batteries. Maybe the costs would be lower if storage was pumped hydro.
1000 km of grid interconnections is clearly inadequate for solar, where potential power providers are literally half a world away much of the time. Wind, too, is usually in synch over areas above 1000km in diameter. Wind farms across the whole of northern Europe are positively correlated. Going as far away as Spain gives a weakly negative link; southern Russia, Turkia and north Africa are other possibilities, assuming that 1) the donor region will have enough excess capacity to keep both ends of the link powered at night, and 2) you trust the governments of said region not to pull a Putin, and cut supply any time it gives political advantage. Cutting off power would cause instant chaos, worse than a fuel supply, where there's more time to react. Japan, South Korea and Taiwan would never allow dependence on wind and solar power from China.
Proponents of '100% wind, solar, water' (usually effectively 95%+ just wind and solar) claim that comparatively tiny amounts of storage will suffice, if a massively overbuilt wind and solar capacity is coupled with a 'copper-plate' grid, where power produced anywhere is effortlessly transported to wherever it's needed. Both those preconditions are far from being met anywhere. California has enough wind and solar to theoretically cover its peak daily demand, Germany has twice as much (151 GW for a peak demand yesterday of 73GW.) Germany still has to complete the HVDC cables that would let it transport North Sea wind power to industry in Bavaria, let alone from Moroccan solar. The North Sea-Bavaria link might be completed by the mid-30s, and will supply, weather permitting, about half as much power at peak as the three reactors Germany closed a year ago did all the time. California last month got 0.29% of its power from grid scale batteries, despite having the world's largest capacity by far. I was a supporter of the proposed Lake Onslow storage scheme in New Zealand, but most countries do not have the topography for projects large enough to cover their power needs for a significant period.
The size of the existing nuclear fleet shows that, at least in the past, it did make economic sense. The number of countries committing to extensive new build - Japan and South Korea, China, India, Egypt, Turkia, most of eastern Europe- suggest that it still does. Ontario has by far the lowest-carbon electricity in North America, apart from in three other Canadian provinces, and parts of Washington State, which are pretty much all hydro. They've recently announced a major refurbishment of four large reactors at Pickering, plans to build four smaller 300MW reactors at the Darlington site, and intentions for many more large reactors, including at Bruce, already the largest operating nuclear plant in the world. Sweden, which correspondingly has the lowest-carbon power in Europe, also announced a goal of two new gigawatt-scale reactors by 2035, and up to ten by 2045. Czechia, with a population of 10 million, has just opened tenders for four gigawatt-scale reactors, to supplement the four GW worth which already provide nearly half its power.
John Oneill at 352:
You are simply making up unsupported BS. You provide no citations to support your wild assertions.
Energy researchers have shown that it is possible to make renewable grids as small as all the individual states in the USA. It is cheaper to make larger grids. The most recent research, like Jacobson et al, make grids that cover most of each continent. Like all North America for example. Your repeatedly mentioning wind in Wyoming is simply BS. No-one says that Wyoming should be able to economically generate all of its electricity 100% of the time using wind alone except nuclear supporters. South Korea, Taiwan and Japan can build out completely renewable systems without relying on China if they choose to. It will be more expensive for them. That is just a political call.
Your claim that Jacobson supports "massively overbuilt wind and solar capacity is coupled with a 'copper-plate' grid" is deliberately false. The overbuild of the grid is less than 10%. The current fossil and nuclear grid has way more excess capacity than the proposed new renewable grids.
I note that nuclear supporters invariably ignore the immense storage needed to have a primarily nuclear grid since nuclear cannot load follow or provide peak power. The needed storage greatly exceeds the storage needed for renewable energy. All the big pumped hydro in the USA was originally built to store excess nuclear power generated at night.
Giving examples like the current California grid do not show that renewable energy cannot power the entire economy. Renewable power has only been the cheapest for about 8 years. There has not been enough time to build out the new renewable grid yet. Battery storage has only been economic for 2 or 3 years. You noting that California only gets 0.29% of power from a brand new technology is simply blowing crap on the discussion. I note that in the past 20 years nuclear power in California has dramatically decreased while battery storage has increased. The future trend is clear.
I note that the current grid took over 100 years to build. Nuclear power is 70 years old and only generates less than 5% of world energy. As an aside there is not enough uranium to generate more than 5% of all world power, the remaining 95% will have to be renewable, mostly wind and solar.
So what if Germany has to increase the size of their grid? They will save money and the environment in the end.
I have followed nuclear power for over 50 years. Nuclear has never been economic. 100% of the existing reactors in the world were built with extensive government subsidies. That is why construction of new reactors in the West has been mostly stopped. Most of the current reactors under construction are being built by the Chinese or the Russian governments.
Please provide one example of a peer reviewed proposed future world power system that uses more nuclear than the currently built reactors.
Whenever I examine nuclear supporters claims closely I find that they are not supported by the data.
Nuclear is not economic, takes too long to build and there is not enough uranium.
Worth noting that the Lake Onslow scheme in NZ was for "dry-year" storage because of NZ high dependence on hydro. Solar and wind are still minor components of its electricity mix and easily backed by ordinary hydro in "normal" years. Ordinary non-pumped hydro also works as storage - hold back water when wind is blowing or the sun shining; release it to create power when its not.
'Please provide one example of a peer reviewed proposed future world power system that uses more nuclear than the currently built reactors.'
'The IPCC found that, on average, the pathways for the 1.5°C scenario require nuclear energy to reach 1 160 gigawatts of electricity by 2050, up from 394 gigawatts in 2020. 1 160 GW by 2050 is an ambitious target for nuclear energy, but it is not beyond reach.' https://www.oecd-nea.org/upload/docs/application/pdf/2021-10/nuclear_energy_and_climate_change_-_cop26_flyer.pdf
Insisting on peer reviewed papers is a good way of ensuring that I can't read them. Here's a calculation on the cost of moving the United States to 100% nuclear electricity. As Joris van Dorp writes, it's a thought experiment on the cost of providing percentages of nuclear from 0 to 100, at various interest rates and build costs. '..It looks like solar and wind are today cheap enough to allow them to work economically as a fuel saving technology with natural gas. And if nuclear costs stay as high as they are today, it even looks as though a combination of storage, wind, solar, demand response and nuclear may be an optimal mix for a zero carbon energy system. However, this does not detract from the fact that nuclear power as a single technological concept is evidently sufficient to allow achieving a low-cost zero-carbon energy system, with no help needed at all from any wind power, solar power or anything else, which is the only thing this article was intended for.'
https://medium.com/generation-atomic/how-much-would-a-100-nuclear-energy-system-cost-3dd7703dd5d3
[PS] Links activated. You have been around long enough to know that you need to create links yourself in the links editor.
John ONeill:
You link to a Nuclear Energy Agency report, hardly an unbiased source, that contains no data or analysis. They link to an IPCC report where the summary for policy makers alone is 24 pages long. The report is hundreds of pages. You must provide a link to an evidence based report and give me the pages that relate to the topic we are discussing. Most proposed future energy systems have a little nuclear since plants currently under construction will presumably still be running in 26 years.
Your other link, which I have previously debunked upthread, is a web piece by a completely uninformed person who has no education or experience in nuclear energy and learned everyting they know about nuclear from the internet. (If you read the rest of this thread you would stop repeating the mistakes nuclear supporters have made upthread). He models the current electrical supply in the USA. Since all cars and all heating by heat pumps will be electrical it is expected that electrical consumption in the USA will at least double. His system is much too small. He uses fossil gas for storage since the required storage would be too expensive to build. I note that a system using fossil gas for storage does not stop emitting CO2 as required. Duh! The cost is prohibitive, he assesses cost incorrectly. The ignorant errors in this analysis are too numerous to address. The fact that nuclear supporters cite this blog proves that nuclear is not economic.
Do you really want to run Afganistan and Yemen completely on nuclear? A solution that does not work for most of the world is hardly a reasonable proposal. There is only enough uranium in all known deposits to run the entire world for 5 years. (Abbott 2012). Read Abott 2012 (linked in the op).
Here is free link to the Jacobson et al 2022 paper titled "Low-cost solutions to global warming,air pollution, and energy insecurity for145 countries". Note that Jacobson describes a solution suitable for the entire world and not just the USA. Upthread I have provided at least a dozen links to free papers that describe completely renewable systems to generate all energy for the entire world. If you read Jacobson you will have more knowledge of what we are talking about. You currently are not very informed.
Here is a free link to a paper titled "On the History and Future of 100% Renewable Energy Systems Research", one of the 59 papers that have cited the Jacobson paper. If you read it you will be more informed about what energy researchers think about future energy systems and make fewer ignorant statements online.
If you have not put in the work to learn how to find papers that support your position it is not my problem. It is not my job to spoon feed you information that you cannot be bothered to read yourself. You have to do your homework if you want to tell other people what they should do. Uninformed proposals do not help advance the discussion.
The fact that you cannot find anything to support your position demonstrates that the nuclear discussion on line is completely fantasy based and not fact based. If documents supporting the nuclear position existed than nuclear supporters would cite them. Nuclear supporters cite industry propaganda as if it were fact based information.
Whenever I examine nuclear supporters claims closely I find that they are not supported by the data.
Nuclear is not economic, takes too long to build and there is not enough uranium.
'Nuclear is not economic' - the 17 countries building new nuclear missed your memo.
'..takes too long to build..' Mean construction time was 7.5 years, with a long tail. Countries involved in a concerted buildout do rather better - Japan averaged less than 5 years, China and South Korea less than 6. Sheffield Forgemasters, one of the few companies qualified to make reactor pressure vessels, has just demonstrated a new method of ion beam welding, letting them weld around the girth of an RPV ring in one day. This weld, on a 4 metre diameter, 200 mm thick piece, with very tight inspection requirements, would normally take up to a year. RPVs have been one of the bottlenecks for nuclear growth. Other solutions, such as the heavy water reactors used in India, don't have RPVs.
'..there is not enough uranium.' This was the perceived reality when the industry was just starting up - and when Cold War bomb-making led to a frantic search for uranium reserves, since enriching to 90% U235 bomb-grade uses up far more feedstock than does the 3-5% used in light-water reactors, or the natural uranium used in mainly Canadian and Indian heavy water reactors. At the time, it was also assumed that energy demand would keep growing at 1960s rates, and that most of the growth would be from nuclear. L Ron Hubbard's famous graph of human energy use rising sharply from a low base, as fossil fuel reserves are used up, and dropping equally sharply back to pre-industrial levels, was used by Peak Oil doomers to predict a coming crash, to be followed by unending scarcity. In fact, Hubbard original graph showed nuclear growing as fast as fossil fuel energy, completely replacing it, and then maintaining that level indefinitely. Plans were in place to switch to fast reactors, converting the 99.3% U238 of natural uranium to fissile plutonium, and to use thorium, 3x more abundant again, as fissile U233. This effort stalled when demand fell, and uranium proved to be much more abundant than thought. Until recently, global production has been well below demand, due to oversupply causing very low prices. Many high grade mines, like MacArthur River in Saskatchewan, were closed during the drop in demand after Fukushima, with the word's third and fourth largest users, Japan and Germany, temporarily shutting their whole industries. With demand now booming, these mines are reopening, and new prospecting has resumed. (Many nuclear operators are on long-term contracts, and have existing stocks, so are not immediately affected.)
Hubbard's fossil peak has been slower to arrive than expected, and so has the nuclear growth he expected to replace it. Long term though, I expect his insight to be accurate. The drive for increasing energy use is still there - nobody wants to stay poor (religious orders aside). The down-ramp on fossil use will be steeper than the rise, as climate concerns spread. Can weather-based energy fill the gap? Not judging by the view out my window (mid summer, 8/8ths cloud cover, national wind fleet at 1/3 of capacity).
I've read some of Mark Jacobson's papers - all the way back to his cover article on Scientific American, in 2009. Before him, there was Amory Lovins' vision of a 'soft path' energy future, very influential on Jimmy Carter's policy. The two were actually diametrically opposite in their prescriptions. Lovins decried the cost and energy waste of the transmission grid, calling for efficiency ('negawatts'), small-scale, local wind and solar, backed by fluidised bed coal. Jacobson wants a maximal grid, moving greatly overbuilt wind and solar across continents, with probably battery backup, no biofuels or combustion energy, no new hydro. Neither prescription has done well when put into practice in reducing emissions. US CO2 emissions per capita hardly changed from the 70s to the 2000s, only falling with the switch from coal to gas (though increased methane leakage may have negated some of the climate benefit). Widespread, government-sponsored wind and solar growth, most notably in Germany, has bought a rapid rise in installation, but though the individual solar plants and wind turbines became much cheaper, their integration into the grid led to increasing power costs, while fossil fuel use persisted at a higher level than on grids that had already switched to nuclear for largely economic reasons.
Some countries whose governments had declared that nuclear power would cease have reversed course, and plan new build - notably Japan, South Korea, Sweden, and Italy. Others - Germany, Spain, Switzerland, Taiwan, which had 20 to 40% of their power from nuclear - currently persist in de-nuclearising. Russia is building plants in Turkey, Egypt, Iran, India, Bangla Desh, and shortly Hungary. Russia, United Arab Emirates, Iran, and possibly soon Saudi Arabia, are building nuclear plants at home because it displaces gas, which earns much more money as exports. Japan and South Korea are building nuclear for the opposite reason - it makes power much more cheaply than imported liquefied natural gas, at East Asian prices. The important question for the future is whether nuclear can take more than a toehold share in countries like India, Pakistan, South Africa, and Indonesia, where energy use is rising fast, and coal is now the chosen option.
John Oneill at 357:
I note that you have made another post without a single cite to suport your wild claims.
Brandolini's Law certainly pertains to this exchange so I will be as brief as possible.
"Nuclear is not economic": All of the reactors currently being built are financed almost entirely by governments. The market has completely rejected nuclear power because it is not economic.
"Takes too long to build": According to the World Nuclear Industry Status report 2023 "For the 58 reactors being built, an average of 6 years has passed since construction start—slightly lower than the mid-2022 average of 6.8 years—and many remain far from completion." while "The mean time from construction start to grid connection for the seven reactors started up in 2022 was nine years," (my emphasis) This includes only construction time. The additional planning time, time to obtain construction permits etc is many years. Typical timeframes for nuclear are 10-15 years. By contrast, wind and solar projects typically take 2-4 years from proposal to completion.
"There is not enough uranium": According to Abbott (2012) as of 2012 there is only enough uranium in known deposits to power the world for 5 years. Nuclear supporters would not be attempting to obtain uranium from the ocean if there was enough uranium on land. You provide no references to support your wild claim that enough uranium exists. Frankly, this is common knowledge among informed people.
Your comments on renewable power are contradicted by experience. Educated readers here will not be fooled. Obviously in the 70's to the 2000's renewable sources did not contribute much because they were not economic at that time. Now they are the cheapest power in the world and are reducing carbon emissions more every day.
According to the World Nuclear Industry Status report, at least Italy, Japan and Sweden currently have no plans to build new reactors. Bertolini's Law applies, I have not checked the rest of your list. I note that France's much heralded announcement about building 6 new reactors will not replace their current 56 reactors that are at the end of their useful life. I note that over 50% of Frances nuclear fleet was offline in the past few years for unplanned repairs due to age. In addition, no money has been budgeted to build the announced reactors.
Meanwhile, according to the IEA:
"Over the coming five years, several renewable energy milestones are expected to be achieved:
In 2024, wind and solar PV together generate more electricity than hydropower.
In 2025, renewables surpass coal to become the largest source of electricity generation.
Wind and solar PV each surpass nuclear electricity generation in 2025 and 2026 respectively.
In 2028, renewable energy sources account for over 42% of global electricity generation, with the share of wind and solar PV doubling to 25%."
I note that the IEA has historically severely underestimated the amount of renewable energy that would be constructed in the future.
Whenever I examine nuclear supporters claims closely I find that they are not supported by the data.
Nuclear is not economic, takes too long to build and there is not enough uranium.
[PS] Removed the "chromeextension" frontspeice from the link. Note that you have to be careful copying links to PDF from the URL bar in chrome.
I noticed that the link to the World Nuclear Industry Status report 2023 will not open on my computer. Here it is:
https://www.worldnuclearreport.org/IMG/pdf/wnisr2023-v4-hr.pdf
you might need to download the report from here:
https://www.worldnuclearreport.org/-World-Nuclear-Industry-Status-Report-2023-.html
[PS] Links activated. Please, please do this yourself.
John ONeill is certainly persistent.
Up-page he has noted that Ontario, Canada is one of the places announcing plans for future nuclear power production expansion. Ontario recently announced plans to refurbish an existing reactor at its Pickering location. This is one of Canada's earliest nuclear power locations. It is not a new plant, although Ontario is also planning to expand nuclear generation facilities operated by Bruce Power (elsewhere in the province).
The Pickering refurbishment is expected to take 11 years, and has already been in the planning/consideration stages for several more. Hardly an example of John ONeill's claim that new plants can be brought online in less than 10 years. The Pickering plant already has many of the needed approvals, since it is an existing reactor complex.
The government of Ontario was short on budget details when making the announcement, but a similar refurbishment nearby had a budget of $12.8 billion. That refurbishment started 7+ years ago and is still not completed. Bruce Power is also in the middle of refurbishment projects that are taking many years.
...and there is considerable debate as to what this will do to power costs. This will not be a cheap process. As Michael Sweet points out, it takes government support to make this happen.
Details in this news story.
John, what is needed to build a station is investors. To get investors, you need rock-solid economic proposition. We are still not seeing that from you. The Lazard report on LCOE noted that costs for continuing a nuclear plant that had capital cost already written off was very low. (ie no economic case for shutting them down). I would guess that reburbishment costs are similarly economic in many cases.
What I certainly don't buy in any proposition that governments should build/pay for nuclear stations because of dearth of investors to do it privately. There are good reasons that investors are going elsewhere.
The grids with the lowest emissions worldwide are not those with the deregulated power markets that came in first under Pinochet in Chile, then in the UK under Thatcher, and subsequently mostly from about 1995 -2004 in parts of the US, Europe and in Australia (1998 under Max Bradford in New Zealand.) Those have not demonstrated the capability to grow capacity, and slash emissions, that will be needed if we want a real transition. For most of the last forty years, in first world countries, demand was static, and the market system only acted to replace existing capacity, not cover expanding uses. (Nor did it give the lowered prices promised when the reforms were introduced.) The lowest emissions are mostly in areas with large, government-run entities, using either hydro - Hydro Quebec and BC Hydro in Canada, Electrobras in Brazil, Statkraft in Norway - or hydro and nuclear - Electricite de France, Ontario Power Generation, Vattenfall in Sweden. Governments can supply the 'patient money', from superannuation funds, needed for long-term energy projects. Private companies, with management usually only staying in the job a few years, pay more attention to quarterly earnings than is needed for projects that will span many generations. A reactor takes years, and a lot of work, to bring online, but once done nothing compares for abundant clean power.