The leading disclaimer: I do not believe that technology in and of itself is the solution to pretty much anything – especially when the technologies discussed don’t even exist at scale yet. Having said that, some back-of-the-envelope math can inform us of the art of the possible and, as such, be useful in guiding us. Finally, nobody, including myself, have double-checked my math here, so yell out if you see any order-of-magnitude errors.
Could airports generate their own fuel from only on-site resources? It sounds like a silly question – but is it?
Shem Malmquist shared a Wired article about turning airports into giant solar farms, and it made me think – what if, indeed, we turned airports into giant solar farms, but what if we weren’t content just generating electricity?
What if we wanted fuel? What if we used that electricity to generate clean hydrogen, on-site, for powering the next generation of aircraft? What if you used local rainwater for that?
Could an airport self-generate all the fuel it needed for its flights? Let’s take Melbourne Airport as an example and do the math.
How much electricity could the airport generate?
Melbourne airport covers an area of 23.69km2. Let’s say we could usefully cover 30% of it with solar panels, giving us a bit over 7km2 of panels.
Given the Melbourne solar insolation values (annual average of 4.43kWh/m2/day) and a 20% solar panel efficiency, that would give us total electricity production of approx. 5,600 MWh per day, or a bit over 2 million megawatt-hours per year.
How much hydrogen could we generate?
Hydrogen electrolysis takes almost 50kWh to produce a kilogram of hydrogen from water, so we could produce approx. 20,000kg with one MWh – or, given our electricity supply of two million MWh, our total production capacity from the electricity point of view would be over 40 billion kilograms of hydrogen per year.
But to produce hydrogen, we need water – would we have enough just from rainwater?
How much water would we have?
Melbourne Airport receives an average of 530mm of rainfall annually. Let’s aim to capture 75% of that run-off.
That would give us 9,400 ML of usable water from the airport area per year.
Is it enough?
Electrolysis produces a bit over 100 grams of hydrogen from every kilogram of water. Our 9,400 ML of water would therefore allow us to produce slightly over one billion kilograms of hydrogen per year.
This is far less than our electricity supply would allow us to do, so we’re water-limited in this scenario. Assuming we don’t want to transport water or electricity off-site either way, we could get away with a much smaller solar installation.
How much would we need though? The energy content of hydrogen – per kilogram – is actually very good, over 100 MJ/kg – much more than the 43 MJ/kg of Jet A-1 fuel. However, the energy density of hydrogen is very poor compared to jet fuel, which creates storage space problems for long-haul flights in particular.
Let’s sidestep that challenge for now, and take Airbus’ word that they can produce their ZEROe hydrogen-combustion-powered aircraft with 2,000nm range by 2035. That’s enough range for all domestic Australian operations.
Anyway, do we have enough?
Pre-pandemic, Melbourne Airport got over 600 flights per day on average. Let’s say the fuel demand of a single trip is, on average, 10,000kg of Jet A-1. A trip to Sydney, the most frequented leg, consumes around half of that, so that’s probably a reasonable ballpark.
That implies a total fuel need of some 6 million kg of fuel per day, or over 2 billion kg per year. Slightly less than half of that weight in hydrogen would have the same energy content, so we’d need around 985,500,000 kg of hydrogen per year to power all the flights from Melbourne.
Slightly under one billion kilograms – which aligns quite nicely with our production capacity of slightly over one billion kilograms of hydrogen per year.
The conclusion
Melbourne Airport could be self-sufficient when it comes not just to its own energy use, but the energy use of all the flights that depart from it.
All the required fuel could, in theory, be generated on-site.
There is more than enough electricity production potential at Melbourne Airport than they would need to power all their flights. We could get away with a much, much smaller installation than we imagined at the start, and still have the energy to spare for direct recharging of all the electric aircraft.
The trailing disclaimer: The propulsion landscape is likely to trifurcate and I kind of omitted the 3rd, most difficult, category of long-haul here. By 2040-2050, we may well end up using electric aircraft for short hops, hydrogen for medium-haul and drop-in biofuels or synthetics for long-haul while using traditional gas turbine engines. And the traditional oil-derived fuels will be with us for decades, too.
I am fully aware that reality is much more complicated than this simple thought exercise. Nevertheless, I think it’s an interesting thought experiment, and it shows that the scenario is not actually as fanciful as I imagined it might be.