Labor says it will reduce 2005 CO2 emissions of 560 million tonnes by 43 per cent come 2030. That brings it down to 57 per cent of 560 = 320 million tonnes. Emissions for 2021 were about 490 million tonnes so it will be reduced by 490 – 320 = 170 million tonnes.
Where does the 43 per cent come from? My guess is the person who selected it was a fan of the late Douglas Adams’ The Hitchhiker’s Guide to the Galaxy, and through a slip of memory thought the answer was 43. The government has now set this target in legislation. Court challenges by emissions-reduction activists to coal or gas expansions, to name a couple, will now have in their armoury this legislation to support their case. The halving of global investment in fossil fuels since 2015 has much to do with the current world energy crisis.
The biggest emitter of CO2 is power generation, at 34 per cent, and that amounted to about 170 million tonnes in 2020. Where is Labor going to find its 170 million tonnes? I don’t know, but if it gets 170 million tonnes of it from power generation, let’s look at the consequences of going 100 per cent renewable at some point in the near future.
If only 80 per cent of coal and gas generation were to be replaced and, importantly, appropriately sized battery back-up is built to support the removed generation, then the cost will be 80 per cent of what I have calculated. If a photovoltaic solar farm of 1 megawatt nameplate capacity ( MW- a million watts) is built, and generates a MW for each hour of the day for a year, it would generate 8760 MWh over 12 months. By monitoring the actual output over a year and adding up all the MWh, to take account of the 4 seasons, it is found for the typical farm that the yearly total is 1750 MWh. This is 20 per cent, or 0.2 times, the 8760 MWh total. This 0.2 is called the “capacity factor” for the PV solar farms. For the average day, the 1 MW farm produces 1752/365=4.8 MWh. Expressed another way – 24 hours x 0.2 = 4.8 hours it outputs 1 MW. In summer it will beat the average and in a wet winter it will be less than the average.
Now let’s look at the 2880 MW NSW Eraring coal fired power station, which is marked for closures in about three years time. If in a 24-hour day it delivers power as following:
♦ 2700MW for the periods 6am to 9am and 5pm to 9 pm ie 7 hours for 18,900MWh.
♦ 2200MW from 9am to 5 pm ie 8 hours for 17600 MWh.
♦ 1700MW for 9pm to 6am ie 9 hours for 15,300MWh.
This gives 51,800 MWh for the 24-hour period. If this is to be replaced with a solar PV farm, how big does it have to be to produce to match this output in 4.8 hours?
Solar farm output x 4.8 = 51,800 MWh, so farm output = 10,800MW. This is 3.7 times bigger than the 2880 MW plant it is replacing. The 4.8 hours is for an average day. In summer more power will be made than required. In winter there will be brown outs.
In the 9am to 5 pm period the demand is 2200 MW, so 4.8 x 2200 = 10,560 MWh of generated power is consumed. The 51,800 – 10,560 = 41,240 MWh has to be stored in batteries if CO2 emissions from coal and gas turbines are disallowed.
The Eraring announcement, reported in the press, said a 700 MW battery was planned as part of the shift to renewable energy. It did not state for how long this battery could output 700 MW — a typical and annoying failure of reporters to supply the MWh value for batteries. Typically, the Eraring battery could do so for 1.3 hours i.e. 910 MWh versus the 41,240MWh required. This 700MW battery is only good for managing transient variations in supply. Where the 41,240 MWh comes from was not identified. Snowy2.0 perhaps?
Greens leader Adam Bandt clearly does not understand the difference between a firming battery and a transient-response battery because he said “when SA needed to firm it’s supply, it built the worlds biggest battery in under a month”. This Tesla battery was 100MW/130MWh, cost $90 million, and if called upon to firm, the loss of a 1300MW supply, it could do so for six minutes.
If a wind farm were used instead of a solar farm, the typical capacity factor is found to be 0.4 and the calculation becomes wind farm output x 9.6 hour = 51,800MWh. (24 x .4 = 9.6). So the farm output = 5400MW, which is 1.9 times the 2880MW coal power station.
To look at the costs involved, I will use a 50/50 mix of wind and solar farms. They are each going to provide 51,800/2 = 25,900MWh. So the solar farm is 25,900/4.8 = 5400MW and the wind farm is 25,900/9.6 = 2700MW. A total of 8100 MW. The current cost of the solar is $1.7 million/ MW and the land based wind farm is $2 million/MW. So cost comes to $1.7 million x 5400 + $2 million x 2700 equals $14.58 billion.
The battery-storage requirement is complicated because who can know when and at what strength the wind will blow. So I will minimise the storage requirement by saying no wind is blowing when the solar farm is operating. If it did, all the wind output would have to be stored because the solar farm by itself could supply the demand.
So the solar farm outputs 4.8 x 5400 = 25,920 MWh and, as shown above,10,560 MWh is consumed during this time, leaving 15360 MWh to be stored. The wind farm outputs between 6am and 9am and 6pm and 9pm, too good to be true, and 3.6 hours during the night. During those periods the demand is 6 x 2700 + 3.6 x 1700 = 22,320 MWh requiring 25,900 – 22,320 = 3600 MWh be stored. The total to be stored is 3600+15,360 = 18,960MWh or 18.96 GWh (G = billion). The battery efficiency is appropriately 90 per cent, so the batteries have to be rated at 18.96/.9 = 21 GWh. Battery storage costs $500 million per GWh, so the battery cost is $10.5 billion.
Managing the haphazard energy supply from renewables falls upon the power- and process-control engineers. For 21 GWh of storage they may install four assemblies each rated at 3GW/5GWh. When the renewables supply more than the demand, the excess is used to charge a not fully charged assembly and when the renewables is less than demand, an assembly supplies all the demand and all the renewable energy is used for charging. This brings the cost to $25 billion and the expense of the transmission lines to convey the electricity from the remote solar and wind locations to where it can be connected to a major transmission line, has to be added onto this figure.
$25 billion/2880 MW = $8.7 million/MW replaced is a sobering number.
Coincidentally, WA coal and gas generation is very close to the output of the Eraring power station. So WA has to add 5400 MW of solar and 2700 MW of wind plus 21 GWh of batteries costing $25 billion to go 100 per cent renewable. Add to this the cost of additional transmission lines.
Alinta intends to replace Loy Yang B 1000 MW coal station with an offshore wind farm and pumped hydro. As Jeff Dimery, Alinta CEO, told The Financial Review’s Energy and Climate Summit in October 2022, “What cost me $1 billion to acquire is going to cost me $8 billion to replace, so let’s talk about that and explain to me how energy prices still come down. I am missing something”.
I don’t have a price on a transmission line cost/km, but Labor’s Chris Bowen sees an urgent need to spend $20 billion on major transmission lines on the National Energy Grid for the eastern states plus SA. Include $4 billion for Snowy 2.0, and an extra 10,000 km of minor transmission lines required to replace coal and gas generation with renewables and $4 million/km is probably a fair ballpark figure.
The National Energy Grid has approximately 35,000MW of coal plus gas generation capacity. WA, which is not part of the National Energy Grid, will increase this amount by about 2500 MW, to 37,500MW. So, for the whole country, we are looking at 37,500/2880 = 13 Erarings. And 13 x $25 billion = $325 billion, plus the cost of transmission lines for 105 GW of wind and solar, 2.8 times the 37.5 GW and 270 GWh of batteries and pumped hydro.
The above calculation indicates 2.8 times more renewables MWs replacing the coal and gas MWs. To suggest that, when renewables are generating more than the grid demands, this is the time to divert that excess to making green hydrogen by electrolysis, doesn’t hold water. If you want to make green hydrogen under my scenario, start adding more GWs to the 105GW total.
As the level of renewables climbs, I expect the unreliability of wind and solar will have become so obvious that the shutting down of gas-fired generation will be suspended. Gas generates 400g of CO2/kWh versus 1000g of CO2/kWh from coal. (I acknowledge my numbers don’t factor in the growth in energy consumption over time). We currently have installed nameplate capacities for wind of 9100 MW, solar farms 7000 MW and hydro 8500 MW. Roof-top solar does not figure in the solar farm number.
Making and storing hydrogen as a way to support the intermittent energy from renewables is not an attractive solution. Elon Musk has dismissed hydrogen as “the most dumb thing I could possibly imagine for energy storage”. Musk also makes the odd dumb statement, but that isn’t one of them. If a GWh of renewable electricity is stored in a battery, it will release only 0.9 GWh when called on. If the GWh was used to make hydrogen by electrolysis it would, at 80 per cent efficiency, make hydrogen with an energy content of 0.8 GWh. This has to be stored somewhere, which also uses energy, until required. If this was used to drive an open circuit gas turbine with 44% conversion efficiency, the electricity generated would amount to 0.8 x 0.44 = 0.35 GWh versus 0.9 GWh from the battery.
Making green hydrogen for ammonia production, industrial heating, hydrogen fuel cell usage and green steel are not to be put in the same basket as electricity production use.
France emits very minor amounts of CO2 when generating its electricity because its power source is 70 per cent nuclear and 19 per cent renewables. They reprocess their spent fuel to produce a mixed oxide fuel (plutonium and uranium) for sending back to the reactor and encapsulate the fission products in glass for long term disposal. This mitigates the issue of spent fuel disposal but it will not satisfy those who are frightened of the nuclear power industry.
The French pay about 2/3 of what the pro-green and anti-nuclear Germans pay per kWh. France has not had a Three Mile Island, Chernobyl or Fukushima. They are adding to their number of operating plants with a new, larger, more fuel efficient design, although this initial model is proving an enormous headache, with cost and time overruns. Critics of nuclear in Australia, such as Chis Bowen, point to this problem the French are having as a reason not to entertain nuclear power as a solution.
If the French shared Bowen’s attitude they would not have 56 operating reactors. To be consistent, Bowen should be opposed to pumped hydro, given the Snowy 2.0 fiasco he inherited. This mad scheme, championed by Malcolm Turnbull as a $2 billion no-brainer, now looks like costing $6 billion-plus, with $4 billion for the transmission line to take the power from where it is generated to where it is to be used.
Bowen does not mention the United Arab Emirates (UAE) embrace of nuclear power. Starting in 2012, it commissioned 4 x 1400 MW Korean KEPCO pressured-water reactors, The first unit came on line in 2021, the second in 2022, the third is near completion and fourth is 90 per cent complete. And the cost comes out at $A7.3 million/MW ($US4.5 million/MW).
In Australia it is probable a move in parliament to overturn our legislative ban on nuclear power would succeed, this expectation based on surveys of voter sentiment. I am not, however, proposing we go for the big, high-pressure light-water reactors (2200 psi) because they face so many regulatory impositions on the design and construction of the reactor and containment vessels and take large amounts of time and money to build. The reactor containment vessel has to withstand being hit by a jet liner. It is so large because it has to contain the reactor cooling water when converted to steam (at about 80 psi) as a result of reactor-vessel rupture as would likely occur in the event of a core meltdown.
Modular high-pressure light-water reactors have been powering nuclear powered submarines for decades. The Virginia class reactor is rated at 210 MW. This class is one of the options under ARKUS. I am waiting on the development of hopefully successfully designed, tested and competitively priced low pressure — like liquid sodium or molten salt cooled, small modular reactors (SMRs) – notionally less than 300 MWe. Without light water for moderation they require fuel enriched to between 5 to 20 per cent U235 (HALEU- high assay low enriched uranium). Joe Biden has allocated $700 million to kick start US production of HALEU. Buying it from Russia was the plan before Feb/2022, and the loss of this source has caused TerraPower to announce a minimum two-year delay in their start-up date for their SMR. These SMR — like the Finnish design above — may enter the US market by the end of this decade. It is proposed these SMRs be placed below ground level where a jet liner can’t hit them, so the enormous 1-2 metre thick containment vessel is not built.
Following commissioning of TerraPower’s first sodium-cooled SMR, subsequent units are projected to have an overnight construction cost of between $US 2.8 and 3 million/MW.
The US NuScale modular light water cooled, high pressure reactor, rated at 70MWe, is further advanced than the above low pressure reactors and maybe it will obtains a full Nuclear Regulator Commission (NRC) licence shortly. It is planning to come online in 2029, located at the Idaho National Laboratory. Using light water as a moderator, it is fuelled with 3- to-5 per cent enriched U235. The US DOE has estimated the plant cost will exceed $US6.8 million/MW. NuScale have estimated the electricity will cost $US58/MWh.
Canada’s Ontario Power Generation plans to install a GE Hitachi BWRX-300 MW SMR and bring it online in 2028. It is predicted to reduce capital cost by 60 per cent per MW compared to big conventional nuclear plants.
If Generation 4 Small Modular Reactors were around now we could begin to replace 37,500MW of coal and gas generation from nuclear plants located beside those retired plants, thus allowing the use of existing cooling towers and transmission lines. If the cost of SMRs were reduced by 50 per cent, to $8 million/MW, the 37,500 MW would cost $300 billion i.e. cheaper and longer lasting than Labor’s option.
BA ‘Tim’ O’Brien is a retired metallurgist