Energy (Part 2: Nuclear)
TLDR: While renewables should be the focus of energy production, nuclear energy is a good baseline source, far less dangerous than fossil fuel and far better than most think it is. New nuclear technologies promise even safer, more efficient ways of generating power.
Prerequsites: Part 1
"There is no likelihood that man can ever tap the power of the atom. The glib supposition of utilizing atomic energy when our coal has run out is a completely unscientific Utopian dream, a childish bug-a-boo."
— Robert Millikan, 1923 Physics Nobel Prize Winner (emphasis added)
Nuclear energy has an extremely bad rap. Nuclear power is strongly associated with proliferating weapons of mass destruction, toxic waste that lasts hundreds of thousands of years, and high-profile disasters like Chernobyl and Fukushima. With such things on the table, it’s understandable that most people think poorly of nuclear energy.
But these fears are often out-of-touch with reality, exaggerating the risks. The truth is that nuclear power can be safe and effective. If we want to curb the pollution from fossil fuels many parts of the world may need to fall back on nuclear power as part of the broader solution to climate change.
Nuclear Proliferation
The best criticism of nuclear energy, in my eyes, is that it has tended to go hand-in-hand with nations developing nuclear bombs. Nuclear weapons of mass destruction are some of the most terrible artifacts ever created. There are good arguments that their advent brought about a new era of peace, but as long as they exist we collectively stand under a nuclear Sword of Damocles, waiting for apocalypse. Many times in history we’ve come near the brink of armageddon. Surely it would be more ideal to find peace without having to place the fate of the world on the knife’s edge.
In 1954, Canada (and the USA) collaborated with the newly-independent country of India to set up CIRUS, a research-reactor near Mumbai meant to help India develop nuclear energy. The reactor was explicitly marked to be used only for peaceful purposes. Twenty years later in 1974, India demonstrated their first nuclear weapons, thanks in part to weapons-grade plutonium produced at CIRUS.
This is the story of dual-use nuclear technology. The same materials used for nuclear energy production can be refined and put towards nuclear weapons. Even when a nation has demonstrated good-will and cooperation with nuclear non-proliferation, such as Iran before 1979, the winds of fate can change and previously benign reactors might become sources of planetary risk.
There are reasonable arguments that Thorium-based reactors (which we’ll talk about more in a bit) are less susceptible to being converted into nuclear enrichment facilities because they don’t produce plutonium. But they still produce enough uranium-233 to make weapons of mass destruction, so aren’t a panacea.
The real solution to preventing nuclear proliferation is the simple, boring path of surveillance by trusted authorities. The world is currently imperiled by biological weapon and artificial intelligence research, both of which are easy to do by small labs. In contrast, nuclear weapons development requires large facilities with clear supply chains that can be detected and outlawed.
And this mostly works! Almost all of the world’s nuclear weapons were developed by nations that were in no way claiming to just be pursuing energy. The USA, Russia, the UK, France, China, Israel, and arguably even North Korea all developed nuclear technology as part of standard weapons research, not dual-use scheming. Only South Africa, India, and Pakistan have betrayed international trust and successfully developed nuclear weapons under the guise of energy research. (Iran and North Korea are edge-cases.) Meanwhile, 21 other countries have had working nuclear reactors without violating international agreements and developing nuclear weapons.
Nuclear Waste
In some circles it’s anathema to call nuclear a “clean” energy technology. Opponents of nuclear power point out that while the CO₂ produced by fossil fuels is bad, at least it’s a common atmospheric molecule that is taken up by plants and can absorb into the ocean. Nuclear waste, on the other hand, is extremely toxic to all life and can last for hundreds of thousands of years.
When a reactor runs, the fuel rods/pellets change from the base metal (usually uranium) into a random assortment of other elements, almost all of which are metals. The specific byproducts depend on the fuel used, but typically include elements clustered around Krypton or Barium. (Aside: Nuclear waste is just boring metal, not glowing green goo. It doesn’t even visibly glow, though it can cause nearby materials to glow, such as the pools that are used to shield it immediately after it comes out of the reactor.)
But occasionally a fuel atom like uranium will absorb incoming radiation without immediately fissioning. Through this process transuranic elements like plutonium can form — their unstable nuclei just waiting to release energy. The slow release of radiation by transuranic elements is what makes spent nuclear fuel dangerous, and it’s through the careful isolation of these elements that nuclear bombs can be made with the spent fuel.
But much as nuclear bombs can be made by reprocessing spent nuclear fuel, one can also separate out the transuranic elements to make… more fuel! France already does this, and the principles of recycling nuclear fuel are well understood. Better yet, by isolating and re-using spent fuel, the most dangerous and long-lived metals are removed, making the resulting substance only dangerous on the order of centuries, rather than hundreds of thousands of years. The main obstacle to recycling is spending money doing the large-scale chemical processing of highly radioactive material, which does indeed sound expensive.
But let’s imagine the worst-case default scenario, faced by many nuclear plants around the world: highly radioactive waste that needs to be stored for geologic time without entering the biosphere. What can be done?
Well, we can just put it deep underground. Seriously, it’s actually not that complicated. Nuclear waste is extremely compact, and can be stored easily and safely in sealed tanks of steel and concrete. Most nuclear plants just store their fuel on-site in what are called “dry casks.”
While storing the waste is a challenge, it’s mostly a political one rather than an issue of engineering. Finland is launching the world’s first commercial nuclear storage facility in 2023.
Centralized storage is a much better idea than letting dry casks sit all across the world. By concentrating the waste in the most stable places it reduces the risk of accidents and theft. We need more facilities like the one in Finland, and to find political solutions so that people will tolerate their proactive construction, rather than waiting for disaster.
Nuclear Accidents
Speaking of accidents, let’s turn now to what is definitely the most attention-grabbing property of nuclear energy: the threat of nuclear meltdown.
As touched on before, 32 nations have collectively operated 438 commercial nuclear power plants around the world. The first nuclear power plant started in 1951, and around one tenth of the world’s electricity comes from nuclear fission. All this discounts nuclear submarines and other military or scientific applications. In all that time there have been two level-7 disasters (the worst possible), one level-6, and about half a dozen disasters for each of the levels below that. Total deaths from all nuclear disasters in history are widely disputed, but my sources indicate that between two hundred and four thousand people have died from nuclear accidents.
In the context of the 70 years and hundreds of nuclear plants, the nuclear death toll is genuinely tiny. This is especially true in contrast with other sources of energy production. Fossil fuels, especially coal, kill millions of people each year.
The Chernobyl meltdown in 1986 was the worst nuclear disaster in all of history. The ineptness involved was staggering and it has been a good case study for nuclear engineers all around the world in how not to design a system, and especially in how to better train and guide the operators of a nuclear plant. (Nearly everything that happened was the result of human error.) In a major way, a disaster like Chernobyl cannot happen again — we are simply more careful now. Best estimates indicate around 400 people died from the Chernobyl disaster, with ~$70 billion dollars in damages (inflation adjusted).
What about Fukushima, the other level-7 disaster? The Fukushima meltdown was the result of the most devastating earthquake in Japanese history (and the tsunami that followed) and the fourth worst earthquake in world history. Around 20,000 people died from the natural disaster, with a total cost estimated at ~$235 billion dollars — the most expensive natural disaster in history. Approximately one person died from the nuclear meltdown. Follow-up studies have shown no detectable increases in long-term cancer rates due to the fallout. Almost all of the deaths and harm at Fukushima were due to the evacuation of the area and the poor ability of the Japanese government to provide elderly and sick evacuees the support they needed.
Readers in the USA may also feel like it’s worth considering Three Mile Island, a 1979 nuclear accident in Pennsylvania, and the worst nuclear accident in America. As a pre-Chernobyl plant, Three Mile Island was designed poorly and its operators didn’t respond appropriately to a basic glitch in the system. As a result of this human error, the plant had a meltdown just twelve days after the blockbuster film The China Syndrome was released, wherein there is a coverup of a nuclear meltdown in a power plant outside Los Angeles. Needless to say, life imitating art caused a huge uproar in the American people, and Three Mile Island was all over the news for weeks. Nobody died. Background radiation levels for those in the nearby town were not significantly elevated. My favorite explanation of the event is this tech talk (37 min).
These three events were all disasters, don’t get me wrong. Many lives have been lost, and billions of dollars of damage have occurred. In working with nuclear energy it is vital to have proper safety as a top priority. But also, we should be careful not to over-update on a few salient disasters while other forms of power generation fill the lungs of innocent people with lethal particulates. If the world had doubled-down on nuclear energy in the 80s instead of scaling back in response to Chernobyl and Three Mile Island, hundreds millions of lives would have been saved, and global-warming would be much less of a problem.
Thorium
No internet discussion of nuclear power would be complete without mentioning thorium-based nuclear power. My favorite source here is whatisnuclear.com, and I strongly recommend reading their thorium explainer, rather than my watered-down version.
In short, thorium is a metal that can be used in place of uranium as fuel for nuclear reactors. The physics is slightly different, and results in a different space of reactor designs. The first (experimental) thorium reactor went online in the 60s, but thorium reactor designs fell by the wayside in decades since then, largely for economic reasons. However, in 2018 the Chinese began construction of a new thorium reactor in Gansu province, near the Gobi desert, which just finished testing and was cleared for start-up. Other Chinese thorium projects, along with designs being pushed by companies in a few other countries, looks to finally kick-start the thorium revolution.
Why is thorium a big deal? Well, largely because it’s well suited to being the fuel for a molten-salt reactor (MSR). In an MSR the fuel (thorium or uranium) is dissolved in a molten salt, and uses that salt to cool and stabilize the reaction. The result is theoretically high efficiency, while reducing the generator’s footprint and vastly improving the safety. Unlike most water-based reactors, MSRs aren’t held under high pressure, which basically eliminates the possibility of an explosion. Furthermore, simple mechanisms can be used to ensure that if the reactor overheats, it can take itself offline, absorbing all excess heat and then cooling down without any human intervention. Thorium MSRs can use more of their fuel than reactors that use solid-metal, and have fewer transuranic byproducts, making the waste less dangerous and reducing the risk of nuclear proliferation.
Thorium reactors are a largely untested technology, but I’m very excited to see what the Chinese and others manage to pull off in the coming years.
Nuclear vs “Renewables”
I’m always a bit miffed at the use of the term “renewable energy.” Solar power, for instance, isn’t really renewable. The sun is a giant nuclear (fusion) reactor, with a limited amount of fuel. Geothermal energy will slowly run out, as it has done on Mars. Rainfall and wind are powered by sunlight hitting the Earth. There is a fixed amount of exergy in our environment, and we have the opportunity to collect and use that energy as the universe slowly decays.
That bit of language aside, where should our attention be, when it comes to producing more energy? Should we spin up more nuclear plants, or build more solar-farms?
We really ought to do both. Global warming is a big deal, and the particulate pollution from fossil fuels ought to be unacceptable even if there were no long-term impacts of the planet’s climate. Nuclear power and other green power sources can work together to make a better world.
On the current margin, I think solar and wind are more exciting, and deserve more support than nuclear energy. The costs of nuclear plants has ballooned over time, especially in the West, where they stopped getting made outside of a couple countries and where regulatory bloat has made them even less profitable. The regulations aren’t entirely unreasonable, either. While most people overestimate the damage from nuclear disasters, meltdowns do happen and they’re quite bad. Furthermore, I think most governments under-invest in solutions for long-term waste storage, and we need stricter global protection against nuclear proliferation.
Nuclear is an exciting field, but not as exciting as the exponential curves seen in the solar/wind space. Given the relatively sluggish speed at which nuclear facilities are built (outside of China, at least), I think doubling-down on solar and wind is the best bet for the near future.
That said, we should be investing in nuclear energy right now, especially in researching faster and cheaper ways of building safe plants. In the winter, when the wind stops, we’ll likely be in much better shape with some kind of energy backstop, and I’d much rather go with a nuclear plant than a coal plant.
Tune in for Part 3, where I’ll talk about Utopia and how to correctly tax carbon emissions.
