MYTH: Nuclear power is dangerous.

The risk of accidents in nuclear power plants is very low and declining. Over six decades, which represents almost 18,000 cumulative reactor-years of operation, nuclear power has remained the safest means of generating electricity per unit energy produced.

There have been three major reactor accidents in the history of civil nuclear power. The first was at Three Mile Island in 1979, which involved a contained meltdown without harm. The second was the well known Chernobyl disaster in 1986 and the most recent was Fukushima in 2011, both releasing radioactive isotopes into the environment to varying extents.

The Chernobyl disaster was caused primarily by an inadequate and poor reactor design that was operated by equally inadequately trained personnel. A steam explosion and subsequent fire resulted in a substantial radiological release into the surrounding environment. Two plant workers died on the night of the accident, and a further 28 people died weeks later from acute radiation poisoning. Another 19 workers died between 1987 and 2004, though their deaths can not be attributed with certainty to radiation exposure. About 6000 cases of thyroid cancer have been observed in the region to date in those that were around during the accident, and fifteen of these cases had proved fatal by 2005 . The United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) says that apart from these increased cases of thyroid cancer “there is no evidence of a major public health impact attributable to radiation exposure 20 years after the accident.”

The Fukushima accident was the result of a major earthquake that induced the reactors into an automatic shutdown. The shutdown triggered backup diesel generators to drive pumps that were intended to circulate the cooling water inside the reactor in the absence of the main power source. The system failed when a 15 metre tsunami created by the earthquake breeched the 10 metre seawall flooding the entire site causing the malfunction of generators. The loss of circulation of cooling water led to a core meltdown in all three reactors. There was an observed significant radiological release into the environment. Despite this, there were no deaths or serious injuries due to radioactivity, though 19,000 were killed by the tsunami and fatalities were recorded due to evacuation of the area.

Apart from Chernobyl, no nuclear workers or members of the public have ever died as a result of exposure to radiation due to a commercial nuclear reactor incident. This is why when measured by fatalities per unit energy produced, nuclear power is by far the safest form of energy, as illustrated in Figure 1.

Furthermore, due to the installed generation of nuclear power displacing coal-fired plants, it has been estimated that this has prevented at least 1.8 million air pollution-related deaths globally to 2013. This number is expected rise to 7 million by mid-century, dependant on what energy sources nuclear power displaces.

Figure 1: Nuclear power is responsible for the least number of fatalities per unit energy produced

If the ban on nuclear power in Australia was lifted tomorrow, and a reactor was to be immediately, it is likely a Generation III+ pressurised water reactor would be chosen. These designs use both active and passive safety system to prevent all of the types of nuclear reactor accidents that have happened in the past. Otherwise, perhaps near-term deployable small modular reactors might be selected for the Australian grid. These reactors enhance safety simply by having a smaller core and thus smaller decay heat that needs removing after shutdown. However, given the likely time frame for implementation Australia may well end up embracing one of the Generation IV reactor designs, which are arguably orders of magnitude safer than the already very safe Gen III+ designs. They use non-water based coolants to operate with inherent safety and use accident tolerant fuels of high melting temperatures or coolants which do not boil even above 800°C.  Head over to the New Nuclear section to learn more about these designs.

MYTH: There is a huge, unsolved nuclear waste problem.

The term ‘nuclear waste’ drives public sentiment for prohibition. This requires a reframing of the definition of this term. It is important to note waste produced by modern reactors is not considered waste- it is more correctly called ‘used fuel’ for the current reactor cycle, or ‘future fuel’ for more advanced reactors. About 20 tons of used fuel is removed from a modern 1 GW light water reactor each year, which is made up of about 96% uranium; 1% plutonium, and the rest fission products and minor actinides.

This used fuel once removed from a reactor still generates a substantial decay heat, so the fuel assemblies are kept in storage pools (onsite near the reactor) for a couple of years. They are then transferred to dry cask storage, as pictured in Figure14 (or in some countries, reprocessing facilities are used to recover the fissile uranium and plutonium). The casks are incredibly robust (watch one withstand a missile strike here).

Figure 2: Cask storage hall, Zwilag storage facility, Switzerland. IMAGE:

If all the used fuel from all the power reactors in the world that have ever operated was stacked on the area of a rugby pitch, it would be about nine metres high. As a comparison, coal plants around the world produce the same volume of waste every hour . This attests to the incredible energy density of nuclear power – this relatively small amount of used fuel over 60+ years has produced a staggering 84,500 TWh of reliable, low-carbon electricity to the world.

During this time, which works out to be almost 18,000 cumulative years of reactor operation, the management and disposal of civil nuclear waste has not caused any serious health or environmental problems, nor posed any real risk to the general public. On any measure, this demonstrates the high level of safety associated with used fuel storage.

Most of the used fuel around the world is currently kept in cask storage, though a number of longer-term solutions are being developed. Next year, Finland’s Onkalo deep geological repository is expected to begin encapsulating casks of used fuel assemblies about 500m below the earth’s surface – the world’s first long-term repository for used nuclear fuel. Recent developments in deep bore disposal techniques have also been shown to be viable solutions for long term storage.

There are many critics of geological storage of used fuel, given the incredible value of the uranium and plutonium it still contains. With uranium prices currently cheap, there is a good argument to keep the used fuel stored safely above ground until it becomes economic to implement reprocessing techniques such as pyroprocessing. During this process, fission products are separated from the useful actinides, and the latter can then be used as fuel in fast-spectrum reactors. The remaining fission products (the only real “waste”) make up a small proportion (3-5%) of the original used fuel, and only have to be stored for 300 years before they have a radioactivity comparable with background levels.

Russia already has two fast-spectrum reactors operating and producing electricity to the grid, and the US, France, Korea, China and India all have designs set for near- to mid-term deployment. The ability for these reactors to consume not only used fuel, but also weapons stockpiles and depleted uranium tailings makes the sustainability case for nuclear even stronger.

It should always be remembered that alternatives for power generation are not without challenges, and their undesirable by-products are generally not well controlled. Solar panels, for example, aren’t easily recyclable, and contain a variety of toxic materials, including lead and cadmium. There is no real plan to ensure this waste stream is kept out of the biosphere, and in the coming years as installed panels get to their end of the life, this is set to a be a pressing problem not just here in Australia, but around the world.

MYTH: Nuclear power is too expensive.

The determination of the real costs of generating electricity from different sources of energy has long been a contentious and largely debatable issue. Predicting what nuclear power might cost in Australia, where the industry is still in its infancy, is difficult. Developing such a new industry from is not without challenges. However, Australia is well placed to observe and utilise best practices from other national nuclear industries and avoid historically observed bad practice.

The cost of building a nuclear power plant varies significantly between countries. There are a few recent examples in the United States and Western Europe that have suffered cost over-runs and construction delays and are often quoted. It is incorrect though to assert that these examples are representative of the entire nuclear power industry. Looking at the broader dataset of historical construction costs gives a better idea on the industry as a whole. Figure 3 plots out the Overnight Construction Costs (OCC) of nuclear power plants expressed in inflation adjusted (2010) US dollars per kW for seven different countries. The OCC is the biggest single cost component of a nuclear plant, and is the most useful and neutral cost metric to compare between countries with differing economies.

Figure 3: Historical dataset of overnight capital costs of nuclear power plants in seven countries

Figure 3 indicates there is a large variation in cost trends across different countries (even with similar nuclear reactor technologies). This variation suggests that other cost factors unrelated to the technology dominated the experience of nuclear power construction and actualised costs within each country. Negative learning curves – where costs increase with generation capacity – can be observed in the United States, West Germany and to some extent Japan, whereas South Korea and India seem to have reduced or at least stabilised costs over time whilst increasing capacity. This suggests that cost factors such as regulatory regime, national utility structure are likely to have significant influence on final cost points. Furthermore India and South Korea tend to have more internationally collaborative industries compared to other nation states, suggesting the competitive advantage of collaboration instead of industry protection.

Figure 4: Projected levelised costs for light water reactors commissioned in 2020

Projected levelised costs of electricity for light water reactors commissioned in 2020 in ten different countries is shown in Figure 4, taken Nuclear Energy Agency’s Projected Costs of Electricity 2015 reportSouth Korea and China are expected to deliver the cheapest electricity from nuclear power, while the US and many other European countries sit around ~US$80/MWh.

Some energy commentators have argued that the Australian grid may be better suited to smaller capacity reactors known as small modular reactors (SMRs) which are expected to have significantly lower capital costs than their larger capacity predecessors. The most mature SMR on the market is from NuScale, whose 12 module 684MWe plant is expected to cost approximately US$3 billion (US$4385/kW). Another SMR design is GE Hitachi’s BWRX-300, which is a 300MWe scaled down version of the mature Gen III+ 1520MWe ESBWR. Projected cost point for nth-of-a-kind commercial deployment is expected to be US$2250/kW.

Figure 5: Projected levelised and capital costs for eight different Gen-IV reactors

Figure 5 is taken from a 2018 study by the Energy Innovation Reform Project, which compiled the submission of individual Gen-IV vendors for commercial (nth-of-a-kind) cost estimates. The capital costs and levelised costs are shown. Across all eight vendors, the average LCOE was US$60/MWh, and average capital cost (including financing) was US$3782/kW.


As these costings for SMRs and Gen-IV designs are provided by the vendors, they should be considered as the minimum cost estimate. It is interesting however to compare these estimates to those that appear in other studies. The two main sources that seem to receive the most attention for Australian pricing projections is the GenCost 2018 report from CSIRO/AEMO and the 2019 AEMO Costs and technical parameter review by GHD. The GenCost report suggests a 2019 capital cost for a large scale light water reactors of just under AU$10,000/kW. This cost is predicted to vary little until 2050. The GHD report quotes a SMR capital cost of AU$16,000/kW for 2018, which also isn’t predicted to improve over a 30 year time frame. These costings and zero learning curve predictions are at odds with the recent construction cost data and in the case for SMRs are as much as four to seven times larger than vendor supplied pricing estimates.

Figure 6: A comparison of system and levelized cost of a range of energy sources in four different countries

To put some of these costs in perspective, it is helpful to compare the cost of electricity generated from nuclear with that of other sources. Levelised cost data taken from the Nuclear Energy Agency’s Projected Costs of Electricity 2015 report along with system costs from their 2019 report The Cost of Decarbonisation is shown in Figure 6 for the main generation sources in four different countries: France, South Korea, the United Kingdom and the United States. In all cases, nuclear power is cost competitive. At a 7% discount rate, South Korea LCOE for nuclear is about US$43/MWh, significantly cheaper than any generating source across all four countries.

What is obvious from this data is that nuclear power can be delivered in an affordable manner. The removal of the prohibition of nuclear power is an important first step in a cost discovery process for the Australian market. The end of the prohibition would provide a market signal to the international nuclear industry which would encourage vendors to develop and present detailed business case to investors for consideration.

For further consideration of cost in an Australian context, the results of modelling undertaken by Electric Power Consulting on Australia’s National Energy Market is discussed under the NEM Modelling section of this site.  It demonstrated that nuclear power can provide a cost-effective decarbonisation solution for the Australian electricity grid.

MYTH: Nuclear power takes too long to build.

A common misconception about nuclear power is that because the construction of nuclear plants have such significant development timescale the technology can’t be relied on to provide a fast enough decarbonisation response to climate change. While there has been construction delays on a number of individual projects, a broader look at the cumulative construction on country by country basis tells a different story.

Many countries have been able to ramp up their generation capacity in nuclear power at faster rates than for solar and wind combined. Figure 6 illustrates this as a function of generation capacity measured in kWh per capita, per year (added annually). This figure shows that nuclear generation in France and Sweden added significant capacity over 10 years in the 1980s. The UAE also added a greater amount of power generation per capita when compared to the biggest renewable energy scale ups in Denmark, Spain & Germany.

Figure 6: Nuclear power can be built fast