To find the best catalyst for green ammonia, researchers were staring down 8000 lab experiments — with AI, they only needed 28.
Scientists and engineers at UNSW Sydney, renowned for their earlier breakthrough in green ammonia synthesis, have harnessed the power of artificial intelligence and machine learning to make the process drastically more efficient.
Ammonia, a nitrogen-rich compound essential for fertilisers, is often credited with averting global famine in the 20th century.
However, its production comes at a steep environmental cost: the conventional Haber-Bosch process demands temperatures above 400 degrees Celsius and pressures over 200 times that of the atmosphere, making ammonia responsible for 2 per cent of worldwide greenhouse gas emissions.
In 2021, a UNSW team developed a method to produce ammonia from air and water using renewable energy at just the temperature of a warm summer’s day.
While this was a leap forward, Dr Ali Jalili from UNSW’s School of Chemistry explains there was room for improvement: could the process be made even more efficient, with less energy waste and higher ammonia yields?
To optimise the process, the team needed to identify the ideal catalyst — a substance that accelerates chemical reactions without being consumed.
As explained in their paper published in Small, the researchers began by shortlisting promising candidates.
“We selected 13 metals that past research said had the qualities we wanted — for example, this metal is good at absorbing Nitrogen, this one is good at absorbing hydrogen and so on,” Dr Jalili stated.
“But the best catalyst would need a combination of these metals, and if you do the maths, that turns out to be more than 8000 different combinations.”
To tackle this daunting task, the team turned to AI. They trained a machine learning system with data on how each metal behaved, allowing it to predict the most promising combinations.
This approach slashed the number of necessary lab experiments from over 8000 to just 28.
“AI drastically reduced discovery time and resources, replacing thousands of trial-and-error experiments,” said Dr Jalili.
“Having a shortlist of 28 different combinations of metals meant we saved a huge amount of time in lab work compared to if we’d had to test all 8000 of them, which was simply not possible.”
The winning formula was a blend of iron, bismuth, nickel, tin, and zinc.
The results exceeded expectations, as Dr Jalili stated: “We achieved a sevenfold improvement in the ammonia production rate and at the same time it was close to 100 per cent efficient, meaning almost all of the electrical energy we needed to make the reaction happen was used to make ammonia — very little was wasted.”
This near-perfect Faradaic efficiency means the process is not only sustainable but also cost-effective and scalable.
Dr Jalili noted that ammonia can now be produced at just 25 degrees Celsius — less than 10 per cent of the temperature required for the traditional Haber-Bosch method.
“This low-temperature, high-efficiency approach makes green ammonia production viable and scalable,” said Dr Jalili.
“We believe it can compete directly with electrified Haber–Bosch and even fossil-based routes, creating a realistic pathway for truly green ammonia.”
The team’s innovation could soon have real-world impact.
Dr Jalili envisions a future where farmers produce ammonia for fertilisers onsite, reducing costs, energy use, and transport emissions.
Localised ammonia production is already in trial phase, with compact, factory-built modules — each the size of a shipping container — available for purchase or lease.
“For a century, ammonia production was based on massive, centralised factories that cut costs by operating at enormous scales, but those projects take years to build, require billions of dollars in capital, and cannot adapt quickly as energy markets change,” Dr Jalili says.
“Our approach breaks away from the era of centralised, giga-scale plants and opens the door for smaller, decentralised units that require much lower upfront investments.”
Low-cost, low-energy ammonia production also supports the transition to a hydrogen economy. Liquid ammonia stores more hydrogen energy than liquid hydrogen, making it a superior candidate for renewable energy storage and transport.
“This same system doubles as a carbon-free hydrogen carrier, creating new economic opportunities that align with the global shift to a clean hydrogen economy,” Dr Jalili said.
Building on their farm-scale proof of nitrogen fertiliser production, Dr Jalili’s team is now deploying their AI-discovered catalyst in distributed ammonia modules to cut costs, sharpen green ammonia’s competitiveness, and accelerate its adoption worldwide.
