Fuel Supply In The Long Term
by David Archibald
1 April 2026
To recap, back in the 1950s it was realised that our fossil fuel endowment would be exhausted one day and nuclear power would have to be commercialised to maintain civilisation at the level to which we have become accustomed. Shell Oil geologist King Hubbert worried about civilisation’s ‘margin of safety’ in getting the right nuclear technology sorted before fossil fuels ran out. There has no real progress in nuclear technology since Congress killed the breeder reactor project in 1983.
Peak oil was expected to happen in 2005. LNG receiving terminals were being built on the US Gulf Coast to import the natural gas that would be needed. Instead, the US shale oil boom started, the LNG receiving terminals were converted into export terminals and the civilisational party went on for another 20 years. The shale bounty was squandered. The oil price is rising again with the peak in US oil production, overprinted by localised shortages due to reliance on supply from the Middle East.
The short-term solution is to build coal liquefaction plants. But we know that our coal endowment will run out, and natural gas before that. Lack of alternatives will force us to the right solution in the end; our only choice is how much pain we endure in the interim. The purpose of this essay is to determine the price points of technology choices on that journey. The common thread is the price of hydrogen. Hydrogen enables the conversion of electrical energy from nuclear reactors to chemical potential.
Carbon is the carrier for hydrogen, so that it can be used in high-energy-density liquid fuels at room temperature and pressure. It is still a wonderful future if we have the wit to seek it out.
There are two choices in coal liquefaction processes: Bergius and Fischer-Tropsch, both invented in Germany in the 1910s. In the Bergius process, hydrogen is forced into coal molecules at a temperature of 450˚C and a pressure of 170 kg/cm2 (165 atmospheres or 2,420 psi). The Fischer-Tropsch process burns coal in pure oxygen to produce a synthesis gas that is catalysed to long chain hydrocarbons in an oil bath. Bergius is the better process. In WW2, German synthetic fuel production was dominantly via the Bergius process:
How much will that cost us? A Bergius plant consuming 10,000 tonnes per day of bituminous coal with 10% ash and 10% moisture will produce 30,000 barrels per day of liquid fuels, plus some LPG. The product split will be:
Petrol 10,500 barrels per day
Jet fuel 4,500 barrels per day
Diesel 15,000 barrels per day
That’s a rough guide; hydrotreating can change long chain molecules into more valuable short chain ones. The cost of the plant is US$3.2 billion, which is A$4.5 billion. Costs on a per litre basis will be:
Coal at US$142/tonne $0.43
Operating costs $0.13
Depreciation over 10 years $0.26
Profit margin $0.39
Total $1.21
Lower grade coal could be used. Ash and water don’t impair the process, they only take up space in the reactor vessel which has a one-hour residence time. The total cost of $1.21 per litre equates to US$135 per barrel, which, after transport and refining costs, is close to the international oil price at the moment.
The wholesale price is affordable. What about the capital cost per consumer? The drive-away price of a Toyota Corolla Hybrid is $37,000. Its fuel consumption rate is 25 kilometres per litre. If the vehicle does the normal 20,000 km per year, that is 800 litres to get there. The capital cost of a litre of annual production is $2.63. If we multiply that by the 800 litres of fuel consumption per annum, the Corolla’s share of the cost of the Bergius plant to supply it is $2,100. This is 5.7% of the drive-away price, less than the cost of a refrigerator or some TVs, and a fraction of the $8,000 you can pay for extra trim for the Corolla. Car buyers should be given the option of buying a perpetual fuel supply for their vehicles.
It is the same story with wheat-growing. Medium-rainfall country in the WA wheatbelt is currently selling for $7,500 per hectare. Each hectare is expected to produce 2.5 tonnes of wheat per hectare, using 15 litres of diesel per tonne in no-till cropping, equating to 38 litres per hectare.
At the moment, that diesel supply is on a hand-to-mouth basis. The farmer might get his crop in, but will there be diesel for sale come harvest? To reduce risk he could buy in the diesel for harvest at the time of planting and keep it in tanks on the farm. Better yet, he could guarantee supply in perpetuity by paying $2.63 per annual litre of production from a Bergius plant for an outlay of $100 per hectare, increasing his capital outlay by 1.3%. The cost of disruption is far, far greater than the outlay for fuel supply to the farm. The same is true for mining, trucking and all the other activities of productive people. And it applies to aircraft:
China has supplied 30% of Australia’s jet fuel consumption. It was stupid to get ourselves into that situation. We could be making all the jet fuel we need ourselves. We wouldn’t be spared the price reaction to Middle Eastern disruption but the money would stay in Australia. The Singapore price for jet fuel is up 167% in the last month:
Everyone else on the planet is now aware of just how stupid Australia has been on this subject. This is a recent headline:
With respect to potential biologically-sourced fuels, some mines in Indonesia have up to 40% palm oil in their diesel tanks. Palm oil solidifies at 33˚C so the tanks are continually stirred to keep the palm oil in suspension. Palm oil sells for the equivalent of about US$1.00 per litre in Malaysia. There is potential to grow palm oil in North Queensland and inland from Derby in Western Australia, once the dams in the Kimberley are built. Palm oil needs two metres of rain or equivalent irrigation per annum. Palm oil can be converted to biodiesel at a cost of about $0.10 per litre.
For self-sufficiency in liquid fuels, we need 33 Bergius plants each producing 30,000 barrels per day at a cost of $4.6 billion per plant, for a total outlay of $152 billion. Somehow we have run up a national debt of $1 trillion in the last 25 years and have nothing to show for it. Building the coal liquefaction plants we need will be an enormous benefit by comparison. We can do it.
Bergius plants are the near-term solution. Longer term it will always be nuclear, specifically lead-cooled or molten-salt cooled breeder reactors (sodium is too messy). The price relationship in getting that nuclear power into a form we can use is:
5 cents/kWh = $2.50/kg of hydrogen = $14/GJ = $84/barrel oil
That is the energy equivalence basis. Conversion losses in each step are likely to make the price of liquid fuel at least 50% higher. The lower we can make the cost of nuclear power, the higher the standard of living of our descendants.
David Archibald is the author of The Anticancer Garden in Australia.