Earth 3: Harnessing the power of the Atom
Nuclear energy is, by far, the most sci-fi energy generation technology out there. Just imagine telling someone from the distant past about the energy sources we currently use (underground chemicals (i.e. fossil fuels), solar energy, wind energy) and then adding: oh and there’s this one technology which works by harnessing the energy inside an atomic nucleus. I can only imagine what this distant ancestor might say!
For that reason, I always find it so strange that the way we currently harness nuclear energy is remarkably similar to how it was done 70 years ago. In fact, the very first prototype used to demonstrate nuclear power was based on the same technology that currently underpins the majority of all nuclear plants today. Currently nuclear power accounts for ~4% of global energy produced (led by the US, China and France) and ~9% of global electricity.
Unlike other clean energy technologies however, nuclear’s contribution to the global energy mix has actually gone down significantly (from a high of ~6% in the early 2000s). This makes it (along with hydropower) one of the only clean energy sources that is playing a shrinking role in our energy mix.
Before we go any further, it is important that we define the two (very) different kinds of nuclear technologies:
Nuclear fission. This is the process by which the nucleus of a large atoms (e.g. uranium, plutonium, thorium) is split into smaller nuclei. This splitting converts a very small amount of mass into energy which is then released. This is the process that all existing nuclear plants use.
Nuclear fusion. This is the process in which small nuclei (e.g. hydrogen, helium, boron) are combined to form larger ones. Similar to fusion, the joining of these two nuclei converts a very small amount of mass into energy which is released. This is the process that powers all stars (including, of course, the sun) and is the process a lot of people are working hard to commercialize.
We’ll focus primarily on nuclear fission as that is the current technology that we use to generate clean energy but we will circle back to fusion at the end.
As always, let’s answer a few questions to help us better understand the potential for this energy technology.
Does nuclear fission have the potential to power the world?
The answer to this is a bit complicated and depends on the nuclear plant considered and the type of nuclear fuel used. If you assume we’re only using the main type of fuel we currently use (Uranium-235 extracted on land) and the kinds of reactors we commonly use (non-breeder reactors) then the answer is that we can only completely power the world using nuclear fission for ~5-6 years before we run out of uranium fuel. If you assume we use all the fuels we have at our disposal and use different kinds of nuclear reactors, then we have enough fuel to power the world for 4 billion years. So the tl;dr here is that we can’t power the world with the way we currently produce nuclear energy but we could if we leverage all of the technologies and fuels we have at our disposal.
Can we make energy cheaply enough from nuclear fission?
Let’s revisit our handy LCOE (or Levelized Cost of Energy) discussed in the last post. The LCOE for nuclear electricity in the US today is ~182 $/MWh. That is markedly higher than the LCOE of competing energy technologies (if you recall wind and solar LCOE were at ~50-60 $/MWh in 2023). There are many theories about why new nuclear power is so expensive in the US today (including a lack of new build over the past 20 years and a challenging regulatory environment) and nuclear costs outside the US can be much lower. In South Korea for example, nuclear costs are currently ~50 $/MWh. So there is no fundamental reason why nuclear power needs to be expensive. Much work needs to be (and is being) done to reduce its cost both in the United States and in many parts of the world (particularly Europe).
In addition to generating electricity from nuclear power, it is also possible to produce heat (though no one is currently doing so). Producing heat from nuclear power will almost certainly be cheaper than producing electricity as all nuclear reactors first produce heat and, if electricity is desired, they then convert that heat to electricity (with significant energy loss).
Do we have access to energy from nuclear whenever we want?
Unlike wind and solar, nuclear power plants can run anytime, day and night. Because their construction cost is a substantial fraction of their total cost (and their fuel costs are relatively low), they are typically run (almost) continuously. For example, in the US the average capacity factor for nuclear power was 93.5% in 2019. That said, in other countries (e.g., France) where nuclear energy is a large percentage of overall electricity mix, nuclear power is ramped up and down to account for overall electricity demand.
Is nuclear energy more dangerous than other forms of energy?
Oftentimes when people think about nuclear energy, they think about famous nuclear disasters including Chernobyl, Three Mile Island and, most recently, Fukushima. This may lead people to assume that nuclear power is dangerous. That is not the case at all! When looking at fatalities per amount of energy generated, nuclear is not more dangerous than solar and wind and 100-1000 times safer than fossil fuels. How is that possible? Well, while there were (unfortunately) fatalities during these nuclear disasters, they have been extremely rare and when looked at holistically, end up being equivalent to the risk of solar and wind energy (where workers risk getting electrocuted or falling off wind turbines high off the ground).
Do all countries and geographies have access to these energy technologies?
No. Not all countries have access to nuclear power with only 32 countries currently operating nuclear power plants. While there are other countries in the process of building nuclear plants (currently Bangladesh, Egypt, and Turkey), that still leaves a very large number of countries that don’t have any access to nuclear power. There are many reasons for the lack of nuclear power adoption, especially in developing countries. These include the fact that nuclear plants today require a large amount of capital to build and produce a large amount of power. While at first glance it might seem beneficial for these plants to produce a lot of power, the challenge is that no country wants one plant to dominate its electricity grid as there will be times it will need to be taken offline (to refuel, maintenance or for unforeseen events). Additionally, developing novel reactor designs is extremely expensive and so nuclear plant designs will need to be licensed and constructed by one of only a handful of potential vendors (many of them state owned).
Ok so what about nuclear fusion?
Nuclear fusion is a very interesting and promising energy technology with billions of dollars going into commercializing different technologies. These investments are going into a variety of different fusion approaches and to both startups and large government projects. What all of these approaches have in common however, is that they are all in the pre-commercialization phase. So far no one has been able to generate more energy from nuclear fusion than the total energy that they needed to put into the system (engineering breakeven).1 Given how early nuclear fusion is in the commercialization stage, it is difficult to predict how much it will cost in the long run.
Conclusion
For the past 60 years nuclear has been an invaluable source of clean energy. If we want it to play an increasingly important role however, we must find ways to significantly reduce its cost and to increasing its deployment.
If the price of nuclear power doesn’t go down what alternatives do we have for baseload power? Perhaps geothermal? Stay tuned for the next post on that energy technology!
Recently, the US National Ignition Facility was able to run a fusion experiment where they released more energy in the fusion reaction than went into it. While this was an important milestone, the overall experiment still consumed more energy than was produced.