Technological breakthroughs

A study published in Nature Food (2025) by Stefano Mingolla and Lorenzo Rosa of the Carnegie Institution for Science provides a comprehensive analysis of the global nitrogen fertilizer supply chain and evaluates pathways for low-carbon ammonia production. The authors argue that transitioning to low-emission production methods is essential for ensuring food security and addressing climate change.

Background

Synthetic nitrogen fertilizers supply nutrients for nearly half of the world’s population. Their key input is ammonia (NH₃), synthesized through the Haber–Bosch process developed in the early 20th century. Between 2000 and 2021, the global population increased by 25%, while nitrogen fertilizer consumption rose by 30%. Demand is projected to continue growing by 2.8–3.7% annually.

In 2020, global ammonia production reached 183 million tonnes, with 70% used in agriculture. Ammonia production consumes around 2% of total global final energy use (8.6 exajoules per year) and emits 450–500 million tonnes of CO₂ equivalent, accounting for 1.3% of total global greenhouse gas emissions. When considering the entire life cycle from production to application, emissions rise to 1.1–1.3 billion tonnes of CO₂ equivalent annually, and up to 7.4 billion tonnes when the entire food system is included — equivalent to 41% of emissions from the food sector.

Production remains heavily dependent on fossil fuels

Each kilogram of ammonia produced consumes 7.7–10.1 kWh of electricity — roughly equivalent to the daily electricity consumption of an average European household. Most of this energy is used for hydrogen production, which accounts for 90–95% of the total energy demand of an ammonia plant.

In 2021, 99% of hydrogen used for ammonia synthesis came from fossil fuels: 70% from steam methane reforming of natural gas, 26% from coal gasification, 3% from electricity, and 1% from oil. The Haber–Bosch process alone consumes 3–5% of total global natural gas production. Water electrolysis powered by renewable energy accounted for only 0.1% of total hydrogen production, despite growing twentyfold between 2020 and 2023.

Average direct CO₂ emissions from ammonia production range from 2.6–2.9 tonnes of CO₂ per tonne of NH₃ when using natural gas, and 3.2–4.5 tonnes of CO₂ per tonne of NH₃ when using coal. China produces 28% of global ammonia capacity (58 million tonnes NH₃ per year), with coal accounting for 85% of domestic feedstock in 2019.

A concentrated and vulnerable supply chain

Globally, there are 407 ammonia production plants, but only 61 countries possess such facilities. Most plants are concentrated in the Northern Hemisphere, while developing countries in the Global South largely depend on imports.

In 2021, 46% of global nitrogen use (49 million tonnes N, valued at USD 39 billion) was traded internationally. Four countries — Russia, China, Qatar, and Saudi Arabia — accounted for more than 40% of total exports. Russia was the world’s largest nitrogen fertilizer exporter in 2021 with 7.2 million tonnes N (16% of global exports). Brazil imported 6.7 million tonnes N (14% of global imports), primarily from Russia.

Of the 3.9 billion people dependent on nitrogen fertilizers for food production, 1.2 billion rely on imported nitrogen fertilizers. The figure rises to 1.8 billion when including those using domestically produced fertilizers made from imported natural gas. Much of Sub-Saharan Africa and Latin America exhibit high vulnerability indices.

A clear example was the energy crisis following Russia’s invasion of Ukraine in 2022: European ammonia production capacity declined by 32% compared with the previous year, with half of that reduction directly affecting the ammonia sector.

Pathways for low-carbon ammonia production

Three large-scale technologies have reached sufficient technical maturity, each involving different trade-offs.

Carbon capture and storage (CCS) combined with natural gas has production costs ranging from USD 201–512 per tonne of NH₃, about 50% higher than conventional methods. This approach reduces emissions by roughly 70% (to 0.6–0.7 tonnes CO₂ per tonne NH₃) and requires less water and land than other alternatives (4.2–13.4 m³ water and 4.6–27.6 m² land per tonne NH₃). However, it still depends on fossil fuels and requires suitable geological formations for long-term CO₂ storage.

Water electrolysis powered by renewable electricity reduces emissions by 65–81% compared with conventional production and eliminates dependence on imported natural gas. Current costs range from USD 337–1,218 per tonne NH₃, two to three times higher than fossil-fuel-based production. Land requirements for solar and wind installations are 20–100 times greater than for CCS systems (72–3,223 m² per tonne NH₃). Electricity demand is also extremely high: 8.6–13.1 MWh per tonne NH₃, requiring gigawatt-scale installations.

Biochemical processes using biomass combined with CCS can potentially achieve negative emissions if biogenic CO₂ is captured and permanently stored. Costs vary from USD 215–1,086 per tonne NH₃ depending on biomass sources. However, land requirements (637–2,894 m² per tonne NH₃) and water consumption (248–4,727 m³ per tonne NH₃) are substantially higher than electrolysis by one and three orders of magnitude, respectively. Biomass feedstock accounts for up to 80% of total production costs.

Decentralized small-scale production

Small-scale decentralized electrolysis plants are being developed as a complementary approach. Located close to demand centers, these facilities reduce transportation costs, emissions, and price volatility. Current production costs range from USD 532–1,894 per tonne NH₃ due to limited economies of scale. Early projections suggest decentralized production could become cost-competitive in suitable regions by 2030.

In Uganda, retail urea prices are currently double production costs because of accumulated expenses from transportation, insurance, tariffs, storage, and logistics. In Minnesota, long-distance transportation also significantly increases fertilizer prices — making such regions promising candidates for decentralized production models.

One technical challenge is handling gaseous ammonia under ambient conditions, which requires specialized storage infrastructure. Currently, only 3% of nitrogen fertilizers are directly applied as ammonia, while the remainder are used in derivative forms such as urea or ammonium nitrate. Several startups are developing aqua ammonia production systems — aqueous ammonia solutions that are less volatile, easier to handle, and safer than anhydrous ammonia.

Policy and outlook

Among 352 announced low-carbon mitigation projects since 2020, only six are currently operational, while 36 are in the investment decision or construction phase, representing around 10 million tonnes NH₃ per year. Most low-carbon production pathways currently involve a “green premium” — higher costs compared with conventional production — posing challenges for low-income countries.

The Inflation Reduction Act provides production tax credits for low-carbon hydrogen through Sections 45V (renewable electrolysis) and 45Q (CCS). Initiatives such as Hydrogen Hubs in the United States and the Hydrogen Bank in Europe are also accelerating the development of low-carbon ammonia projects.

Existing ammonia plants currently have an average operational lifetime of 25 years (European plants average 40 years, compared with 12 years in China). Without decarbonization, existing facilities are projected to emit an additional 15.5 billion tonnes of CO₂ over their remaining operational lifespan.

The study was published in Nature Food, Volume 6, June 2025.
DOI: 10.1038/s43016-025-01125-y