Is nuclear power an option for Australia?
Peter Farley | April 3, 2019
People often ask why Australia does not have commercial nuclear power plants despite mining a lot of uranium. Setting aside the political controversies which have kept nuclear power off the electoral agenda, this is akin to asking why don’t we make semiconductors, when we have a lot of sand to make silicon.
The cost of the refined uranium ore, known as yellowcake, that we export amounts to 10% of the cost of the finished fuel assembly, which in turn is less than 10% of the cost of nuclear power. Uranium ore therefore accounts for less than 1% of the cost of nuclear power.
Our wealth in uranium resources therefore has little relevance to the arguments about whether or not to adopt nuclear power, and a litany of other problems all tend to mitigate against the nuclear option when searching for alternatives to coal.
Lack of flexibility
We need an alternative to coal, and nuclear power stations can provide base load, but they perform poorly when asked to increase or decrease their load in reponse to changes in demand. When newly refuelled, a nuclear power plant can be ramped from 50% to 100% in a matter of hours, but as the fuel ages the plants become less flexible and must run at or near full power. There are methods of reducing power below 60% but these can take days or even a week to restore full power.
Nuclear power plants must therefore be complemented by other forms of energy generation or storage to cope with rapid increases in demand. France and Japan rely heavily on nuclear energy and use hydro and pumped hydro power to cover peaks in demand. Japan has 29GW of pumped hydro in comparison to 53GW of nuclear capacity, for example, while France imports energy from abroad at times of peak demand and exports when demand is lower.
The USA aimed to have enough plants to ensure some were in the flexible zone while limiting nuclear power to less than a quarter of total supply, allowing all plants to run at near capacity even at times of minimum demand. capacity.
None of these options would work in Austrlaia. We have no bordering countries to export to and we don’t have enough hydro capacity to cover peaks in demand. Australia could theoretically support a dozen 1.1GW plants running at 75% capacity, but if two or three of these were offline on a day of high demand then the grid would suffer a full-blown power crisis.
Victoria suffered power cuts when it lost 1,800 MW of coal, the equivilent to less than 2 nuclear stations. The loss of nuclear power plants to maintaince is by no means unprecedented. In 2015 all five of Switzerland’s reactors were offline for many months, while in October 2017, 21 of the 58 French reactors were not operating.
Australia’s greater variations in load create another problem. In the past, the Base Load – an much misunderstood term – was usually 40-50% of the maximum load, but in Victoria and South Australia the minimum load is less than 30% of maximum load and trending lower.
Indeed, the minimum load in South Australia can be just 20% of the maximum. That means that while one or two coal or nuclear plants might run at 80% capacity factor, the average demand across the whole system might only be 45-55%. Hydro and gas plants have run up to a quarter of the year to cover peaks in demand in the past, but if nuclear plants were used, more storage and fast response capacity would be required to cover times of high demand.
Emergency back-up
There is also the question of emergency backup in the short, medium and long term. Short term responses to sudden changes in load are covered by “spinning reserves” of underutilised capacity, running on the grid in case of a generation or transmission failure. A1GW generator needs 1 GW of spare capacity in the system which in practice means 3-4GW of gas, hydro, coal, nuclear even wind or solar running at 35-65% capacity.
Minimum demand in South Australia might be 600MW on a balmy night, meaning the state would have to export 1.5 to 2.4 GW, depending on its back up plants, if it had a 1GW nuclear reactor. There may not be a market for this power in other states, nor is there the interconnector capacity to transfer it.
A system with 15 or 20 gas, coal or hydro generators producing 300 or 500MW each allows other plants to cover shortfalls when one plant encounters a problem. However a nuclear plant takes hours to shut down and many hours to restart, meaning a simple steam leak might take 4 days to repair. Pumped hydro could not cover this shortfall, let alone the six week gap required every three years for refuelling or the six months downtime oa major overhaul. Demand tends to peak in times of hot weather, but the same conditions reduce the efficiency of thermal plants, meaning even more capacity is required.
Nuclear power also needs large amounts of cooling water. A single 1.1 GW nuclear plant, the most common size, needs 20-25 GL of water every year – about 7% of Melbourne’s water supply. Melbourne would need 4 or 5 plants to supply all its energy needs, consuming a third of its fresh water. While plants can be cooled by sea water, more complex pumps and maintainance is required, to clear pipes of sealife.
Long lead times
Then there is the question of time. There are just 3 companies in the west capable of building nuclear reactors, Areva, Toshiba/Westinghouse and Korean KHNP. Areva is in financial difficultes while Toshiba Westinghouse has announced it will not build any new reactors and KHNP is heavily subsidised, obscuring the real costs of its plants.
New nuclear plants take a decade to commission after construction is approved, and changing financial circumstances can doom them to failure. Toshiba and Hitachi have both abandoned new nuclear plants in the UK, for example, after investing billions of pounds when it became uneconomic to continue.
Australia would need three or four years to set up a Nuclear Regulatory Commission, issue licenses and arrange finance for new plants, before they could even start construction. Even if a decision to go nuclear was made, power might not be generated until 2037.
Exorbitant costs
Despite all these drawbacks, the high cost of nuclear power is now the greatest stumbling block to its adoption. Eleven nuclear plants in the USA have closed or are closing because they can’t compete with with gas, wind and solar. State governments are subsidising another 5 to keep them open but additional plants have been abandoned or are way over budget already.
Even if Australia could build nuclear plants at the same cost as elsewhere – despite our lack of experience – the power they generated would cost three times as much as renewable sources, while being less responsive to changes in demand and taking far longer to build. A 1.1 GW nuclear plant would have a Capex of A$18-20 billion, for example, while 8,200GWhr per year from a combination of new wind/ solar PV/solar thermal/ pumped hydro and batteries at today’s prices would cost “just” $6-6.5b.
This combined system would have a summer peak capacity of 1,600 MW vs 980MW for the nuclear plant and a 40-80% response time of 30 seconds compared to 30-90 minutes for a nuclear/ gas combination. Not only that, it could be operational by the end of 2023 – when the first sod might be turned for the first nuclear plant.
The false dawn of new technology
While there are several new technologies which may lower the costs of nuclear power and increase its flexibility, these ideas have been “four to five years away” since the 1970’s. Billions of dollars have been spent on developing Small Modular Nuclear Reactors in the US, UK, China, France, Japan and Russia but while half a dozen test plants are working, they remain far from general commercial adoption.
Thorium has long been suggested as an alternative fuel to uranium, but questions over their life span remain, as would the usual problems over licensing, financing and construction. Even China’s aggressive nuclear program will provide no more than 15% of its power. We need action to replace our coal plants long before any of these solutions can make a meaningful contribution to Australia. Even if they do prove practical, it’s highly unlikely that nuclear plants will ever supply more than 10% of the nation’s electricity.
The renewable alternative
Rapid change in power production is possible without nuclear production. Britain’s reliance on coal dropped from 42% in 2012 to just 5% last year, with no coal used to generate power at all for long periods. In the first 25 days of March this year, Germany generated 14.7 TWh from wind and only 11.1 TWH from all fossil fuels and solar in Germany will produce more power than gas going forward.
Our investment in renewables means that wind and solar will increasingly dominate our energy production. Half of Victoria’s power consumption of 105-110 GWh per day, for example, might be covered by wind generation in the near future, while new rooftop installations and solar production plants will increase capacity from 2 Gwh per day to 15 in 2021.
If wind and solar can 90 Gwh, with hydro and gas producing 15 GWh a piece, then Victoria could go coal free for days at a time without energy imports. Given that Tasmania will be able to deliver 10 GWh/day and SA 12, it is quite possible that by 2021 Victoria could go fossil fuel free for even longer periods.
In practice , this won’t happen because brown coal plants will be kept going to undercut NSW black coal generation while gas and hydro will be saved for low wind/solar days. There will however be days when renewables will supply 70% of Victoria’s needs and overall, by the 2022/23 financial year, renewables will account for up to 45% of Victorian power consumption and more than 35% of generation.
SHARE WITH:
Peter Farley is the President of the Victorian Vernier Society and former Deputy President of the Victorian Committee of Engineers Australia. He has also served as a Director of the Inner Melbourne VET Cluster for more than a decade.