Papex

Open problem · Chemistry · Open

Ammonia synthesis at ambient conditions

Is there a scalable way to make ammonia from nitrogen at room temperature and atmospheric pressure?

Summary

Industrial ammonia comes from the Haber process, which needs high temperature and pressure. Laboratory routes, including electrochemical ones, work under milder conditions, but none has approached the Haber process’s efficiency and ease at scale.

Why it is open

Industrial ammonia comes from the Haber–Bosch process, which combines nitrogen with hydrogen made from methane over a promoted iron or ruthenium catalyst at 300–500 °C and 100–300 atmospheres [10]. About 175 million tonnes are made each year, and Olivia Westhead and colleagues attribute 1% of global energy consumption and 1.4% of global carbon dioxide emissions to the process [10]. Brian Hoffman and colleagues write that nearly half of the human population could not exist without the nitrogen fertilisers it supplies [1]. The triple bond of the nitrogen molecule is very strong, 941 kJ mol⁻¹, yet the enzyme nitrogenase breaks it at room temperature and atmospheric pressure [10].

What is unresolved is whether an artificial system can do the same at useful rates and efficiencies. No single body sets the bar. Westhead and colleagues cite US Department of Energy targets of 90% faradaic efficiency (the share of the electric charge that ends up in ammonia) and 300 mA cm⁻² of current density [10]; Jaecheol Choi and colleagues cite yield-rate targets from 100 to about 930 nmol of ammonia per second per square centimetre of electrode [5].

There is also a problem of evidence. Suzanne Andersen and colleagues write that the amount of ammonia produced is usually so small that it cannot be firmly attributed to nitrogen reduction rather than contamination from air, breath, membranes or nitrogen compounds in the gas or the catalyst [3]. A claim therefore has to show, usually with nitrogen-15, a heavier isotope that marks the gas fed in, that the ammonia came from nitrogen gas [3, 5]. Several of the strongest results below were also obtained at 15–20 bar rather than at atmospheric pressure [7, 8]. In 2020 Choi and colleagues concluded that "the question of the feasibility of the direct electrocatalytic reduction of N₂ to ammonia with aqueous electrolytes remains open" [5].

What is known

Biological nitrogen fixation. Nitrogenase, found only in some microorganisms, has two components: an Fe protein that delivers electrons, each transfer driven by hydrolysis of ATP, the cell's energy carrier, and a MoFe protein whose iron–molybdenum cofactor binds and reduces nitrogen [1]. Each nitrogen molecule reduced is accompanied by one obligatory molecule of hydrogen, which Hoffman and colleagues call an apparent "waste" of two reducing equivalents and four ATP [1]. Westhead and colleagues give the enzyme a faradaic efficiency of 66% at ambient pressure, rising to 75% at 50 atmospheres [10]. Molecular catalysts that imitate the enzyme make ammonia but deactivate quickly: the first Schrock catalyst sustained 6 catalytic turnovers, and the Peters group's iron and cobalt complexes about 13 and 2 [10].

False positives (2014–2020). In 2019 Andersen and colleagues proposed a protocol in which nitrogen-15 gas is cleaned of labile nitrogen compounds and recycled to cut its cost [3]. Following it, they found that "no ammonia is produced when using the most promising pure-metal catalysts for this reaction in aqueous media", while lithium electrodeposition in tetrahydrofuran, an organic solvent, did make ammonia [3]. In 2020 Choi and colleagues checked 127 papers on aqueous nitrogen reduction, published up to April 2020, for a sufficient yield rate, reliable nitrogen-15 experiments and control of oxidised nitrogen; none met all three [5]. In their own tests, bismuth and carbon-supported gold, reported among the most active catalysts, showed "immeasurably low performance" [5]. The isotope gas itself can mislead, since commercial nitrogen-15 can carry labelled nitrogen oxides at up to 0.1 mol% [5]. A 2014 Science report of ammonia made electrochemically from air and steam in molten hydroxide at 200 °C was retracted in 2020: its authors found a trace nitrogen-oxyanion (NOx⁻) impurity in their iron oxide and wrote that this impurity, "rather than N₂, is the major nitrogen reactant" [6].

Lithium-mediated reduction (1993–2026). Westhead and colleagues regard this process, pioneered by Akira Tsuneto and colleagues in 1993 and 1994, as the only electrochemical system so far rigorously verified as an ammonia producer [10]. Lithium ions are reduced to lithium metal, which reacts with nitrogen to form lithium nitride; a proton carrier then releases ammonia and frees the lithium ions [8]. Reports from the 1990s gave faradaic efficiencies of 50–60% at 50 bar of nitrogen, and five to six times lower performance at ambient pressure [8]. In 2021 Bryan Suryanto and colleagues replaced sacrificial proton sources with a phosphonium proton shuttle, reaching 53 ± 1 nmol s⁻¹ cm⁻² at 69 ± 1% faradaic efficiency under 0.5 bar of hydrogen and 19.5 bar of nitrogen [7]. In 2022 Hoang-Long Du and colleagues reported 150 ± 20 nmol s⁻¹ cm⁻² at a current-to-ammonia efficiency close to 100%, with a concentrated imide lithium salt under 15 bar of nitrogen [8]. In 2023 Xianbiao Fu and colleagues coupled nitrogen reduction with hydrogen oxidation in a 25 cm² flow cell and reached, at 1 bar, up to 61 ± 1% faradaic efficiency and 13 ± 1% energy efficiency (the fraction of the electrical energy input stored in the ammonia) [9]. In 2024 Shaofeng Li and colleagues ran such a cell for 300 hours at 1 bar and room temperature with a chain-ether solvent, at 64 ± 1% faradaic efficiency [12]. Fu and colleagues also showed that calcium can replace lithium, verified with the rigorous protocol, at 40 ± 2% faradaic efficiency [11]. In February 2026 Qiang Zhang and colleagues reported, in a single study, 98% faradaic efficiency and 21% energy efficiency at 100 mA cm⁻²; faradaic efficiency stayed at 80% for 40 hours before performance declined [14].

Photocatalysis. In 2017 Hiroaki Hirakawa and colleagues reported ammonia from water and nitrogen on titanium dioxide with surface oxygen vacancies under ultraviolet light, at a solar-to-chemical efficiency of 0.02%, which they called the highest among early photocatalytic systems [2]. Yunxuan Zhao and colleagues examined the usual ammonia assays and warned that, because interfering substances are rarely considered, yields reported in many photocatalytic and electrocatalytic studies "may be exaggerated and erroneous" [4]. In 2024 Daniel Maldonado-Lopez and colleagues showed that nitrogen-containing ligands such as oleylamine, used to make metal oxide nanocrystals, can produce false positives in nitrogen photofixation [13].

Main approaches

Direct electrocatalysis in water. Nitrogen is reduced on a solid catalyst in water, powered by renewable electricity, with the hydrogen supplied by oxidising water [10]. Strongest evidence: the route avoids high temperatures and flammable solvents and can in theory work at reasonable overpotentials [5], the excess voltage needed because of slow kinetics [10]. Main difficulty: no aqueous report met all three of Choi and colleagues' reliability criteria [5], hydrogen evolution competes for the electrons, and in calculations a scaling relation between reaction intermediates fixes the minimum limiting potential of the associative pathway, which nitrogenase probably also follows, at about −0.5 V against the reversible hydrogen electrode [10]. Westhead and colleagues note that aqueous carbon dioxide reduction faces the same competition [10]; see the Papex page Fuels from artificial photosynthesis.

Metal-mediated reduction in organic electrolytes. Lithium or calcium is plated and reacts with nitrogen [8, 11]. Strongest evidence: isotope-verified ammonia [3, 11], faradaic efficiencies near 100% [8, 14] and 300 hours of continuous operation at 1 bar [12]. Main difficulty: energy. Because lithium is plated at about −2.7 V against the reversible hydrogen electrode, a cell coupled to water oxidation could reach at most about 30% energy efficiency even without other losses [10]; reported values are 13% and 21% [9, 14]. Solvent stability is a further problem: tetrahydrofuran polymerises and evaporates [12].

Photocatalysis. Light absorbed by a semiconductor drives the reduction of nitrogen with water [2]. Strongest evidence: ammonia from water and nitrogen at a reported 0.02% solar-to-chemical efficiency [2]. Main difficulty: efficiencies are low [2], and assays and catalyst precursors can produce false positives [4, 13].

Enzymes and molecular catalysts. Nitrogenase shows that nitrogen can be reduced at ambient conditions, and its mechanism guides synthetic design [1, 10]. Strongest evidence: 66% faradaic efficiency at ambient pressure for the enzyme [10]. Main difficulty: the enzyme spends ATP and makes obligatory hydrogen [1], molecular catalysts complete few cycles [10], and Hoffman and colleagues describe plants that fix their own nitrogen and catalysts built on the biological mechanism as goals still unmet in 2014 [1].

Open sub-questions

  1. An aqueous catalyst that passes the checks. Can any solid catalyst in water make nitrogen-15-labelled ammonia, above the background from oxidised nitrogen, at 100 nmol s⁻¹ cm⁻² or more, the lowest target Choi et al. cite [5]?
  2. Mediators other than lithium. Westhead et al. tie the energy limit of the lithium route to the potential at which lithium is plated [10]. Can calcium, which mediates at 40% faradaic efficiency [11], or another metal exceed the 13–21% energy efficiency reported for lithium [9, 14]?
  3. Interphase lifetime. What degrades Zhang et al.'s layered interphase after 40 hours at 100 mA cm⁻², and would the chain-ether solvent behind Li et al.'s 300-hour run extend it [12, 14]?
  4. Photocatalysis under the protocol. Does titanium dioxide with oxygen vacancies still make ammonia when tested with cleaned, recycled nitrogen-15 and nitrogen-free precursors [2, 3, 13]?
  5. The enzyme's hydrogen. If nitrogen binding in nitrogenase requires releasing hydrogen, as the reductive-elimination mechanism proposes, must a synthetic catalyst using the same intermediate also lose a quarter of its electrons [1]?

Work on this in Papex

If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.

Key references

  1. Hoffman, B. M., Lukoyanov, D., Yang, Z.-Y., Dean, D. R. & Seefeldt, L. C. (2014). Mechanism of nitrogen fixation by nitrogenase: the next stage. Chemical Reviews 114(8), 4041–4062. 10.1021/cr400641x · Checked 15 Oct 2026
  2. Hirakawa, H., Hashimoto, M., Shiraishi, Y. & Hirai, T. (2017). Photocatalytic conversion of nitrogen to ammonia with water on surface oxygen vacancies of titanium dioxide. Journal of the American Chemical Society 139(31), 10929–10936. 10.1021/jacs.7b06634 · Checked 15 Oct 2026
  3. Andersen, S. Z., Čolić, V., Yang, S., Schwalbe, J. A., Nielander, A. C., McEnaney, J. M. et al. (2019). A rigorous electrochemical ammonia synthesis protocol with quantitative isotope measurements. Nature 570(7762), 504–508; author correction 574(7777), E5. 10.1038/s41586-019-1260-x · Checked 15 Oct 2026
  4. Zhao, Y., Shi, R., Bian, X., Zhou, C., Zhao, Y., Zhang, S. et al. (2019). Ammonia detection methods in photocatalytic and electrocatalytic experiments: how to improve the reliability of NH₃ production rates? Advanced Science 6(8), 1802109. 10.1002/advs.201802109 · Checked 15 Oct 2026
  5. Choi, J., Suryanto, B. H. R., Wang, D., Du, H.-L., Hodgetts, R. Y., Ferrero Vallana, F. M., MacFarlane, D. R. & Simonov, A. N. (2020). Identification and elimination of false positives in electrochemical nitrogen reduction studies. Nature Communications 11, 5546. 10.1038/s41467-020-19130-z · Checked 15 Oct 2026
  6. Licht, S., Cui, B., Wang, B., Li, F.-F., Lau, J. & Liu, S. (2020). Retraction of "Ammonia synthesis by N₂ and steam electrolysis in molten hydroxide suspensions of nanoscale Fe₂O₃" (Science 345, 637–640, 2014). Science 369(6505), 780. 10.1126/science.abe0412 · Checked 15 Oct 2026
  7. Suryanto, B. H. R., Matuszek, K., Choi, J., Hodgetts, R. Y., Du, H.-L., Bakker, J. M. et al. (2021). Nitrogen reduction to ammonia at high efficiency and rates based on a phosphonium proton shuttle. Science 372(6547), 1187–1191. 10.1126/science.abg2371 · Checked 15 Oct 2026
  8. Du, H.-L., Chatti, M., Hodgetts, R. Y., Cherepanov, P. V., Nguyen, C. K., Matuszek, K., MacFarlane, D. R. & Simonov, A. N. (2022). Electroreduction of nitrogen with almost 100% current-to-ammonia efficiency. Nature 609(7928), 722–727. 10.1038/s41586-022-05108-y · Checked 15 Oct 2026
  9. Fu, X., Pedersen, J. B., Zhou, Y., Saccoccio, M., Li, S., Sažinas, R. et al. (2023). Continuous-flow electrosynthesis of ammonia by nitrogen reduction and hydrogen oxidation. Science 379(6633), 707–712. 10.1126/science.adf4403 · Checked 15 Oct 2026
  10. Westhead, O., Barrio, J., Bagger, A., Murray, J. W., Rossmeisl, J., Titirici, M.-M. et al. (2023). Near ambient N₂ fixation on solid electrodes versus enzymes and homogeneous catalysts. Nature Reviews Chemistry 7(3), 184–201; author correction 7(3), 225. 10.1038/s41570-023-00462-5 · Checked 15 Oct 2026
  11. Fu, X., Niemann, V. A., Zhou, Y., Li, S., Zhang, K., Pedersen, J. B. et al. (2024). Calcium-mediated nitrogen reduction for electrochemical ammonia synthesis. Nature Materials 23(1), 101–107 (published online 2023). 10.1038/s41563-023-01702-1 · Checked 15 Oct 2026
  12. Li, S., Zhou, Y., Fu, X., Pedersen, J. B., Saccoccio, M., Andersen, S. Z. et al. (2024). Long-term continuous ammonia electrosynthesis. Nature 629(8010), 92–97. 10.1038/s41586-024-07276-5 · Checked 15 Oct 2026
  13. Maldonado-Lopez, D., Huang, P.-W., Sanchez-Lievanos, K. R., Jana, G., Mendoza-Cortes, J. L., Knowles, K. E. & Hatzell, M. C. (2025). Nitrogen-containing surface ligands lead to false positives for photofixation of N₂ on metal oxide nanocrystals: an experimental and theoretical study. Advanced Functional Materials 35(3), 2413319 (published online 2024). 10.1002/adfm.202413319 · Checked 15 Oct 2026
  14. Zhang, Q., Li, H., Yu, P., Liu, P., Sun, N., Wang, Y. et al. (2026). Enhanced Li-ion diffusion improves N₂-to-NH₃ current efficiency at 100 mA cm⁻². Science 391(6786), 724–729. 10.1126/science.adw5462 · Checked 15 Oct 2026

Page reviewed 15 Oct 2026

Source

Open, as listed by Wikipedia · Unsolved problems in chemistry. Checked 08 Oct 2026.

The status is the source’s. Papex does not decide whether a problem is solved.

Records on this problem

No public record addresses this problem yet.

More in Chemistry