Ammonia synthesis is one of the most important chemical processes as it sustains global food production, but it is a highly polluting and energy-intensive process. Here, the challenges of decarbonizing the process to synthesize green ammonia are discussed.
You have full access to this article via your institution.
Download PDF
Download PDF
Most of us learnt about the synthesis of ammonia through the so-called Haber–Bosch process in secondary school. Likely because it is one of the chemical reactions with the highest impact in our society (ammonia sustains about 50% of the world population), and also because the Haber–Bosch process is an excellent teaching illustration of where chemistry and engineering come together to activate the triple bnd of the nitrogen molecule and overcome thermodynamic limitations. This over 100-year-old process uses ossil fuels (in most cases, methane) as a source of hydrogen and fuel for the process, making up over 2% of the global energy consumption, thus being responsible for over 2% of global CO2 emissions.The synthesis of ammonia, once at the centre of the chemical revolution, is now again at the centre of the green chemical and energy transition1 (green ammonia leads to green fertilizers but it can also be used as a carbon-free energy carrier). Herein, the challenges of the decarbonization of ammonia synthesis are discussed. This process is a model example of the challenges associated with the use of intermittent and seasonal renewable energy for the electrification of the chemical industry, known as Power-to-X (ref. 2).The conventional Haber–Bosch process is more than the ammonia synthesis reactor. It is a highly integrated process that maximizes the extraction of H2 and energy (in the frm of steam) from fossil fuels, and produces the stoichiometric N2:H2 ratio required for ammonia synthesi (without the need to separate N2 from O2). In addition, the poisoning of the ammonia synthesis catalyst is avoided by the removal of COx impurities.The modern Haber–Bosch process consists of a sequence of unit operations that are broadly similar across all plants3. Methane (the predominant feedstock) and steam are fed into a series of reformers to generate hydrogen and carbon monoxide. The reaction heat is provided by combusting additional methane in a furnace surrounding the first reformer and by combusting part of the hydrogen with air within the inlet of the second reformer. The combustion with air in the second reformer also adds the stoichiometric amount of nitrogen for ammonia synthesis. The carbon monoxide generated by reforming is further reacted with steam to produce hydrogen in the water–gas shift reaction, followed by the removal of carbon dioxide, and the reconversion f oxide impurities back to methane to avoid catalyst poisoning.In the final step, the ammonia synthesis ga containing nitrogen and hydrogen along with residual methane and inert argon is compressed and fed to the Haber–Bosch loop. Within this process, nitrogen and hydrogen react under high pressure and temperature over a catalyst up, with low conversions (<20%) limited by thermodynamics, thereby requiring condensation to remove ammonia before recompressing and feeding unreacted nitrogen and hydrogen back to the reactor.Several technological developments and significant energy integration strategies make the efficiency of the best available technology approach 62–65% (with a maximum thermodynamic limit of ~78%)1. But such integration makes the process particularly inflexible, with plant operators generally reluctant to decrease capacity below ~75% and only at low transition rates (~2% per hour) with safety and integrity concerns associated with process instability affecting overall nergy consumption and reducing the lifetime of reactors, furnaces, catalysts and heat recovery boilers and xchangers.Defossilization of this process to produce green ammonia, by using renewable energy for the production of green hydrogen via water splitting, completely breaks the integration developed over 100 years between the current hydrogen synthesis step (via steam reforming of fossil fuels) and the ammonia synthesis loop. It leads to serious process challenges, mainly associated with the intermittent nature of renewable energy supply (Fig. 1), which is anathema to traditionally steady-state chemical processes. Other process challenges and opportunities are associated with the distributed nature and high capital (as opposed to operating) costs of renewable energy. Economic feasibility demands radical changes in the industry, challenging the well-established concepts of centralized, continuous, 24/7 operation processes, benefiting from the economy of scale. Nevertheless, technologcally, green ammonia production is feasible, as currently shown by the first commercial green ammonia producion facility at Puertollano (Spain) run by Fertiberia.Fig. 1: The intermittent and distributed nature of renewable energy supply is at the centre of the process challenges faced in the synthesis of green ammonia.The foundations of chemical process design, operation, integration and optimization have to be reconsidered, potentially moving away from large-scale, steady-state, 24/7 operation.Full size imageThe intermittent and scattered nature of the renewable energy supply is, without a doubt, the main process challenge for green ammonia synthesis. The combination of different renewable sources (such as wind and solar farms) is an attractive way of reducing, although not eliminating, daily and seasonal variations. In green power-to-ammonia plants, feeds to the processes are inherently intermittent due to the nature of the production of renewable energy, but the ammonia synthesis ractor requires steady-state operation; other parts of the process, such as the electrolyser for hydrogen prodction, can flexibly vary their capacity whilst, as mentioned above, ammonia production requires almost constant capacity (see Fig. 1). Achieving process alignment currently focuses on the use of buffers (such as hydrogen tanks and batteries) to align energy generation with ammonia production. However, buffers can only economically overcome short-term energy variations, and they rapidly increase the overall capital costs of the plant4.Prompted by the expense of accommodating a steady-state ammonia synthesis reactor, novel strategies to provide flexible capacity (for example, ramping capabilities) are being developed by industry and academia. Previous approaches to vary the Haber–Bosch loop capacity (such as varying the recycle flow rate) are slow, and there is potential to lead to a runaway reaction due to the exothermicity of the system. Therefore, the need for flexibility offes an opportunity to redesign and optimize the system using new holistic process designs. Particularly attractie is the replacement of condensation by absorptive ammonia separation, opening the door to medium pressure operation (~30 bar) (ref. 5) with intrinsically higher-capacity flexibility. We have recently redefined the Haber–Bosch loop into a single-vessel, recycle-less process by integration of the synthesis and absorptive separation6,7. However, flexible capacity processes lead to low capital utilization, breaking the founding principle of mass production of commodities and opening the door to modular, small-scale processes — which is harmonious with distributed renewable energy.As the challenges with green ammonia production are being addressed, hybridization of existing ammonia plants can stimulate the transition to net zero, accelerating the green ammonia market and advancing the learning curve on green hydrogen production, while also continuing to utilize current assets. In his scheme, both brown (from fossil fuels, also called grey) and green (from renewable energy) hydrogen are fedto the Haber–Bosch loop. The main difficulties of a hybridized process come from the disruption of the heat integration and the balancing of the N2:H2 ratio. Since a fraction of (green) hydrogen is produced via non-integrated electrolysers, there is comparatively less steam generated through brown hydrogen production to compress the synthesis gas into the Haber–Bosch loop. Similarly, green hydrogen needs to be accompanied by nitrogen. For low hybridization ratios, excess air can be fed to the secondary reformer to maintain the N2:H2 stoichiometry, but this approach uses more fuel and it is limited by the reformer exit temperature. For higher hybridization ratios, nitrogen could be added through a separate air separation unit.The ease of hybridization is dependent on the specific design of the existing plant. In some cases, steam is already being imported to balance the system or alternatively the compressor drives are electric rather than steam powered (such as the Imperial Chemical Inustries Leading Concept Ammonia (ICI LCA) design), thereby nullifying the breakdown of steam integration3. For other well-known process designs (for example, the Braun Purifier, Foster Wheeler AM2, ICI Ammonia V (AMV)), excess air (50–200%) is fed to the secondary reformer to improve process efficiency, with surplus nitrogen removed either prior to the Haber–Bosch loop or through the purge8,9. It is likely that process designs such as these will be more suitable for hybridization with green hydrogen than the conventional steam reforming design, but this depends on the specific equipment constraints when given novel operating parameters. In particular, flexible incorporation of green hydrogen by hybridizing an existing reactor has high economic potential to reduce hydrogen buffer size, but may be severely constrained in practice, generating limits on the maximum and rate of cange of the green-hydrogen:brown-hydrogen ratio. However, it is likely that economic viability is considerably beow technical limits.These process issues will be fully applicable to those ammonia plants currently announced to produce blue ammonia (the same process as brown ammonia with the addition of carbon capture facilities) with the plan to transition to green ammonia in the future, unless bio-derived methane is to be used as means of decarbonization. Careful economic and technological analysis should be carried out to truly assess the long-term viability of implementing blue ammonia production.The future of green ammonia in power-to-ammonia plants fully driven by renewable energy is coupled with the onset of a revolution in the chemical industry, with new ways of designing, operating and financing electrified chemical processes, essential to fulfil our decarbonization ambitions. The finances of the processes change to a capital-intensive model with low operational costs if producion of renewable energy and hydrogen production is considered part of the ammonia synthesis process. Novel optimiztion approaches are required, similar to the ones which are in use to account for the variations in the natural gas price, but with substantial shorter-term variations (hours versus days or weeks) considering renewable energy supply, grid electricity, technological constraints and financial incentives in a dynamic environment. System models, previously unnecessary in the chemical industry, will soon become as important as future technological developments. At the same time, the next decade will witness a burst of research and innovation, including the development of new catalysts with enhanced stability under changing conditions and enhanced activity at low temperatures, new materials with higher resistance to thermal and mechanical stress, new separation processes, and novel heat integration strategies to replace steam. Decarbonization of the chemical industry using renewble electricity — Power-to-X — will rejuvenate the chemical engineering field because it will require a truly multiisciplinary effort that reconsiders the foundations of chemical process design, operation, integration and optimization.
ReferencesSmith, C., Hill, A. K. & Torrente-Murciano, L. Energy Environ. Sci. 13, 331–344 (2022).Article
Google Scholar
Daiyan, R., MacGill, I. & Amal, R. Acs Energy Lett. 5, 3843–3847 (2020).Article
CAS
Google Scholar
Appl, M. in Ullmann’s Encyclopedia of Industrial Chemistry (Wiley-VCH, 2006).Armijo, J. & Philibert, C. Int. J. Hydrog. Energy 45, 1541–1558 (2020).Article
CAS
Google Scholar
Malmali, M., Wei, Y., McCormick, A. & Cussler, E. L. Ind. Eng. Chem. Res. 55, 8922–8932 (2016).Article
CAS
Google Scholar
Smith, C. & Torrente-Murciano, L. Adv. Energy Mater. 11, 2003845 (2021).Article
CAS
Google Scholar
Smith, C. & Torrente-Murciano, L. Chem Catal. 1, 1163–1172 (2021).Article
CAS
Google Scholar
Appl, M. Modern Ammonia Technology: Where Have We GotTo, Where Are We Going? (BCInsight, 1993); https://link.gale.com/apps/doc/A13997662/AONEChristensen, K. G., Gosnell, J. H. & Grotz, B. Nitrogen 181, 31–37 (1989).
Google Scholar
Download referencesAcknowledgementsWe thank Fertiberia for allowing us to visit their green ammonia production plant at Puertollano, Spain, and particularly their COO, David Herrero, and plant director, José Antonio Cabello Granados. We also thank the UK Engineering and Physical Sciences Research Council for funding through projects EP/X031683/1 and EP/X016757/1.Author informationAuthors and AffiliationsDepartment of Chemical Engineering and Biotechnology, University of Cambridge, Cambridge, UKLaura Torrente-Murciano & Collin SmithAuthorsLaura Torrente-MurcianoView author publicationsYou can also search for this authorin
PubMed Google ScholarCollin SmithView author publicationsYou can also search for this author in
PubMed Google ScholarCorresponding authorCorrespondence to
Laura Torrent-Murciano.Ethics declarations
Competing interests
L.T.-M. is currently an independent member of the Fertiberia advisory board. C.S. declares no competing interests.
Rights and permissionsReprints and PermissionsAbout this articleCite this articleTorrente-Murciano, L., Smith, C. Process challenges of green ammonia production.
Nat. Synth (2023). https://doi.org/10.1038/s44160-023-00339-xDownload citationPublished: 16 June 2023DOI: https://doi.org/10.1038/s44160-023-00339-xShare this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard
Provided by the Springer Nature SharedIt content-sharing initiative
