Chinese Researchers Use AI to Turn Wastewater Pollution into Fertilizer, Opening a New Path for Green Agriculture

11

Every year, hundreds of millions of tons of agricultural, industrial, and municipal wastewater carrying high concentrations of nitrate flow into rivers, lakes, and groundwater. These pollutants threaten drinking water supplies, fuel harmful algal blooms, and create oxygen-depleted “dead zones” that devastate aquatic ecosystems.

At the same time, producing the ammonia that underpins modern agriculture remains one of the world’s most energy-intensive industrial processes. More than 90% of global ammonia production still relies on the century-old Haber-Bosch process, which converts nitrogen from the air into ammonia under extremely high temperatures and pressures using natural gas or coal. According to the International Energy Agency (IEA), ammonia production accounts for roughly 2% of global final energy consumption and approximately 1.3% of energy-related carbon emissions.

One problem is environmental pollution. The other is the costly production of an essential industrial chemical. Yet both revolve around the same element nitrogen.

Now, a Chinese research team has demonstrated a promising way to connect these two challenges by directly converting nitrate pollutants in wastewater into ammonia, simultaneously treating contaminated water while producing a valuable chemical feedstock.

The breakthrough was achieved by a research team led by Han Lili at the Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences. Their findings were published on March 18 in the Journal of the American Chemical Society (JACS) and were featured on the journal’s cover. The South China Morning Post later described the work as opening a new avenue for low-energy waste-to-resource technologies.

Unlike conventional single-atom catalysts, dual-atom catalysts consist of two neighboring metal atoms that work cooperatively throughout complex chemical reactions. Each atom can perform complementary functions, facilitating electron transfer, stabilizing reaction intermediates, and precisely controlling the breaking and formation of chemical bonds.

These characteristics make DACs particularly well suited for nitrate reduction, a reaction that involves multiple intermediate steps before ammonia is produced. 

However, designing an effective dual-atom catalyst has long been a major scientific challenge. Hundreds of possible metal combinations exist, and experimentally testing them one by one is both time-consuming and expensive.

To overcome this bottleneck, the Chinese researchers incorporated artificial intelligence into the catalyst design process.

Using a deep learning model, the team rapidly screened numerous metal combinations and predicted which pairs would most likely form stable structures with superior catalytic performance. Only the most promising candidates were then synthesized and experimentally validated, dramatically reducing the development cycle compared with conventional trial-and-error approaches.

Ultimately, the researchers successfully fabricated 14 precisely engineered dual-atom catalysts containing various rare-earth elements, including yttrium, scandium, lanthanum, cerium, samarium, europium, erbium, and ytterbium.

Even more remarkably, these catalysts achieved unprecedented metal loadings ranging from 12.8% to 30.7% by weight more than four times higher than previous benchmarks providing a significantly larger number of active catalytic sites.

The team then evaluated the catalysts in electrochemical nitrate reduction experiments using nitrate-rich wastewater. Credit to the abundance of highly active metal sites on the catalyst surface, ammonia production reached approximately 2.7 times that of conventional catalysts, approaching three times the efficiency of comparable systems while generating fewer unwanted byproducts.

In practical terms, this means that the same volume of wastewater can produce substantially more ammonia while simultaneously removing nitrate pollution more effectively. The reaction also proceeds under ambient conditions through electrochemical reduction, avoiding the extreme temperatures and pressures required by the Haber-Bosch process and potentially reducing overall energy consumption.

For decades, wastewater treatment and fertilizer production have operated as two completely separate industries. Conventional wastewater treatment focuses primarily on removing nitrate before discharging treated water. Although effective, this process consumes significant amounts of energy and operating costs while discarding nitrogen that could otherwise be reused.

Yet nitrate itself is simply another nitrogen-containing compound, and nitrogen is precisely the key ingredient required to produce ammonia, the foundation of virtually all nitrogen fertilizers.

If nitrate pollutants can be directly converted into ammonia, wastewater treatment would no longer be merely a pollution-control process. Instead, it would become a resource recovery system capable of generating economic value while protecting the environment.

This circular approach has the potential to reduce pollution, lower dependence on fossil-fuel-based ammonia production, and improve overall resource efficiency.

The technology could also have broader strategic implications. Global fertilizer markets have become increasingly vulnerable to fluctuations in energy prices and geopolitical tensions. Much of the world’s urea exports originate from the Middle East, where natural gas supplies and shipping routes through the Strait of Hormuz remain susceptible to regional instability.

Disruptions to natural gas supplies often translate directly into higher ammonia and fertilizer prices. Earlier this year, for example, India, the world’s second-largest importer of urea—was reportedly forced to purchase approximately 2.5 million tons of urea at nearly double the price paid just two months earlier.

China has largely maintained stable domestic fertilizer production through coal-based ammonia synthesis. Nevertheless, regardless of whether ammonia is produced from natural gas or coal, conventional synthesis remains highly energy-intensive.

Although recovering ammonia from wastewater is unlikely to replace large-scale fertilizer plants in the foreseeable future, it could become an important complementary source of nitrogen, improving the resilience and security of fertilizer supply during periods of energy market volatility.

Despite its promise, the technology is still at the laboratory stage. The reported experiments were conducted under controlled conditions on a relatively small scale. Real-world wastewater is considerably more complex, containing heavy metals, organic contaminants, suspended solids, and numerous other impurities that may reduce catalyst activity or shorten its operational lifetime.

In addition, several practical challenges remain unresolved, including large-scale catalyst manufacturing, integration with existing wastewater treatment infrastructure, long-term operational stability, and ensuring that the electricity required for electrochemical conversion comes from low-carbon energy sources.

Addressing these engineering and economic issues will be essential before the technology can be deployed commercially. Even so, the study demonstrates an important new direction for sustainable chemistry.

By combining artificial intelligence with advanced catalyst design, Chinese researchers have shown that wastewater pollutants can be transformed into valuable chemical resources rather than simply being removed and discarded.

If successfully scaled up, this waste-to-fertilizer approach could simultaneously reduce water pollution, lower the carbon footprint of fertilizer production, improve nitrogen recycling, and strengthen agricultural sustainability.

Source: Xinhua, guancha, sohu, sina