Technology overview: four pathways to produce green ammonia and auxiliary technologies for an ammonia-based economy

Iris ten Have
Extantia Capital
Published in
19 min readJun 6, 2023

This article is part of our green ammonia series. Read the main green ammonia article here. Keen to know about the market opportunity? See this article.

Green ammonia is expected to take off as a sustainable fuel in the maritime industry. Credits: Chris Pagan on Unsplash.

By Iris ten Have, Fernanda Bartels, and Yair Reem

Our atmosphere is composed of roughly 20% oxygen (O2) and 80% nitrogen (N2) gas. The gaseous form of nitrogen is, however, chemically and biologically unusable. Neither animals nor plants can utilise nitrogen as a nutrient directly. This all changed radically in 1908 when Fritz Haber discovered that ammonia (NH3), a usable form of nitrogen, could be created from nitrogen and hydrogen (H2) using an iron-based catalyst at elevated temperatures (450–600°C) and pressures (100–250 bar). Carl Bosch subsequently developed the process on an industrial scale and therefore the process was dubbed “Haber-Bosch”. Since then, this process has literally changed the world: without synthetic ammonia as fertiliser, we would only have been able to sustain about half of today’s population.

The world population with and without synthetic fertiliser (ammonia) created via the Haber-Bosch process. Credits: Our World in Data.

Additionally, fertilisers are essential for bioenergy and biofuel production. According to the international energy agency (IEA), bioenergy is the largest source of renewable energy globally, accounting for 55% of renewable energy and over 6% of the global energy supply. With the global population as well as bioenergy set to increase, the need for synthetic fertiliser will only continue to grow.

Although feeding a ceaselessly growing global population was the main motivation for inventing the Haber-Bosch process, there is a dark side of the medal that’s often overlooked. The other motivation to produce a reactive form of nitrogen was to provide a raw material for explosives to be used in weapons. As a German patriot, Fritz Haber was enthusiastic about developing explosives and other chemical weapons, which to his mind were more humane, because they “would shorten the war”. The Haber-Bosch process came in handy during the First World War: providing Germany with a home supply of ammonia, as their access to natural sources, such as Chilean saltpetre, was cut off. Synthetic ammonia was used to produce nitric acid, ammonium nitrate, nitroglycerine, trinitrotoluene (also known as TNT), and other nitrogen-containing explosives. Ironically, Haber-Bosch thus also partially fueled the World Wars and has since then become the central foundation of global ammunition and explosive supplies.

It is clear that the Haber-Bosch process can be used for good and for bad. Let’s not forget, however, that over 80% of synthetic ammonia is used as fertiliser and that the remaining 20% goes to a wide variety of products: plastics, fibres, dyes, pharmaceuticals, nitric acid, and explosives. What Fritz Haber and Carl Bosch could perhaps not have foreseen at the beginning of the 20th century are the (unintended) environmental consequences of their process. Per tonne of ammonia produced roughly 2 tonnes of CO2e are emitted directly.

Despite the related greenhouse gas emissions, the ammonia demand has been predicted to increase in the coming decades through fertiliser, hydrogen carrier, and shipping fuel. Keeping the net zero in 2050 scenario in mind, there is an increased interest to find green alternatives to create ammonia. In the sections below, we will touch upon multiple technology options: Haber-Bosch with green hydrogen, direct electrolysis, photocatalytic, and biological pathways. We will also go through the auxiliary technologies required for an ammonia-based economy and will discuss the current status of implementation.

But first: how do we assess and compare the different technologies? This is where key performance indicators (KPIs) come in. Our four core KPIs when it comes to green ammonia production are technology readiness level (TRL), energy input, production costs, and carbon footprint. We will use these KPIs throughout the text to discuss the various technologies.

Haber-Bosch with green hydrogen

Using the traditional Haber-Bosch process and replacing grey hydrogen with green hydrogen seems the easiest solution to produce green ammonia, as the Haber-Bosch process itself remains in essence the same. However, with this pathway, everything depends on the availability of cheap green hydrogen. It might sound simple but is in practice a multi-faceted challenge. Firstly, the energy efficiency of water electrolysis to produce green hydrogen has to improve. Secondly, the general electrolyser costs have to come down. And thirdly, the costs of cheap renewable energy are the main cost driver of green hydrogen and should be as low as possible. In the best-case scenario, it will likely take at least until 2035 for green hydrogen to reach price parity with grey hydrogen. That said, government support schemes, such as subsidies for green hydrogen and carbon pricing could significantly speed up this process.

Let’s briefly touch upon green hydrogen made through water electrolysis. The most commonly used systems are the alkaline, polymer electrolyte membrane (PEM), and solid oxide electrolysers (see Table 1), which can be distinguished by the type of electrolyte employed. Out of these three, the alkaline electrolyser is currently the most common one for green hydrogen production. It has been commercialised due to its high technology readiness level (TRL) and low CAPEX costs thanks to the use of less expensive catalysts. PEM is the second most mature technology and its main characteristics include a solid (polymer) electrolyte, compactness, flexibility to work with intermittent electricity, and high-pressure operation. Although the solid oxide electrolyser has the lower TRL at the moment, it is promising for the future due to its high efficiency.

Another consideration to keep in mind when using green hydrogen for the Haber-Bosch process is heat integration. In the past century, engineers have literally undertaken every single step they could possibly take in terms of process optimisation. In other words: the final boss has been defeated when it comes to optimisation. For example, every little bit of heat generated during the process has been integrated elsewhere.

By utilising the waste heat from the exothermic Haber-Bosch process, the steam methane reforming reaction (used to generate the hydrogen required for the Haber-Bosch process) becomes thermally self-sustaining, reducing the need for external energy sources. This integrated approach maximises the overall energy efficiency of ammonia production, showcasing a remarkable synergy.

Now comes the dark cloud though: when we produce hydrogen with an electrolyser instead of via steam methane reforming, the excess heat produced by the Haber-Bosch reaction cannot be reintegrated in most cases. The operating temperatures of alkaline or PEM electrolysers are simply too low; only a solid oxide electrolyser could waste heat integration potentially be leveraged. If the excess of heat goes to waste instead, the overall process becomes less energy efficient and thus more expensive.

What does it cost?

Regarding the costs involved in green hydrogen production: the costs are initially (in 2020) correlated with TRL according to alkaline < PEM < solid oxide. By 2050, however, that order is predicted to change to solid oxide < alkaline < PEM. This shift is mainly due to solid oxide’s higher system efficiency and higher learning rate, resulting in lower CAPEX and OPEX.

Regarding green ammonia production, the Ammonia Energy Association’s target costs for 2030 are 0.48 USD/kg and for 2050 0.32 USD/kg. Currently, grey ammonia produced through Haber-Bosch costs around 0.33 USD/kg. A carbon tax will likely be an important instrument in bringing down the costs of green ammonia further and faster while driving the costs of grey ammonia up. With carbon taxes, the production prices for green ammonia (in USD per kg) are predicted as follows in 2050: solid oxide: 0.35 < alkaline: 0.42 < PEM: 0.60. These predictions were, however, made with lab-scale data assuming high CAPEX and OPEX and no economies of scale. Another important point to consider is that the main cost driver of ammonia production may change: the cost of natural gas is the main driver for grey ammonia production with hydrogen sourced via methane reforming, while for green ammonia production, the main cost driver will be renewable electricity if the hydrogen is sourced from water electrolysis. All in all, it seems plausible that green ammonia (at least when hydrogen is produced with alkaline or solid oxide water electrolysers) will become cost competitive with grey ammonia by leveraging carbon taxes and economies of scale.

Greenhouse gas emissions

In an ideal world, green ammonia production would be carbon neutral or even carbon negative. In reality, however, our electricity grid has a carbon intensity by itself and therefore (at the moment) green ammonia would still be related to some CO2 emissions. Compared to the traditional Haber-Bosch process, which emits 2.17 kg CO2e/kg NH3 due to grey hydrogen production through steam methane reforming, the electrolyser processes all do better. Alkaline emits 0.83–0.93 kg CO2e/kg NH3, PEM 0.82–0.99kg CO2e/kg NH3, and solid oxide 0.69–0.27 kg CO2e/kg NH3. That said, the electricity grid’s carbon intensity may of course come down over the coming decades.

🚀 Potential advances to be made

Direct electrolysis

The direct electrolysis pathway aims to create green ammonia directly from water and nitrogen from the air. Although this could be very interesting, as the need for green hydrogen is omitted, the technology has a much lower TRL (2–4). The main challenges include that breaking the bonds in a nitrogen molecule costs a lot of energy and is slow. Besides, the nitrogen electrochemical reduction competes with the hydrogen formation reaction. This all results in low ammonia formation rates and thus low efficiencies. Moreover, direct electrolysis is competing against the highly optimised Haber-Bosch process: the energy efficiency of the natural gas-based ammonia production process is as high as 60–70%.

What does it cost?

Although the exact costs of direct electrolysis are difficult to model at the moment, it is clearly an energy-intensive process. As we’ve previously explained in our Chemistry 101 article: breaking chemical bonds costs energy. The main bottleneck is the large amount of energy required to break the triple bond between the two nitrogen atoms (N) in a nitrogen molecule (N2). Therefore, the main cost driver is, like with many green processes, the availability of cheap renewable electricity. This also effectively means that, just like the green Haber-Bosch process, the main cost driver of the ammonia production process would change from the cost of natural gas to the cost of renewable electricity.

Greenhouse gas emissions

Predicting numbers on greenhouse gas emissions for direct electrolysis to make green ammonia is speculative at the moment, but the carbon intensity of the electricity used is inevitably going to be the main driver.

🚀 Potential advances to be made

  • Decreasing the high overpotentials required to overcome the kinetic barrier of N-N cleavage. The kinetic barrier makes the electrochemical process energy intensive. Reducing the energy input could be done through a more effective catalyst, for example. A good catalyst would generally also speed up the reaction rate and improve the time yield.
  • Improving yields and efficiencies (current Faradaic efficiencies (FEs) linger around 10–15%). An issue here is that ammonia formation competes with the hydrogen evolution reaction. Yields should increase at least one order of magnitude for commercially viable projects.
  • Technologies should aim to meet the targets set by the US Department of Energy for commercial electrochemical synthesis of ammonia.
  • Current density: should be 300 mA/cm2
  • Faradaic efficiency: should be 90%
  • Energy efficiency: should be 60%
  • Reaction rates (yields): should be 9.3 x 10–7 mol/cm2/s
  • Improving the limited solubility of nitrogen gas in water. This could be done through advances in the electrolyte or through specific reactor design. Gas diffusion electrodes, for example, facilitate improved contact between nitrogen gas and the electrolyte and could lead to higher current densities.
  • Integrating electrochemical routes with other driving forces, such as elevated temperature (increases reaction speed), elevated pressure (increases nitrogen solubility), plasma (improves faradaic efficiency at low currents, but making plasma requires additional energy), or light.
  • Catalyst longevity: the system has to run for thousands of hours. Right now, partial catalyst deactivation happens in most studies reported in less than 100 hours.

Photocatalysis

Photochemical ammonia production utilises solar energy and a semiconductor material to transform nitrogen and water into ammonia. In essence, water is oxidised by photogenerated holes and nitrogen is reduced by photogenerated electrons yielding ammonia and oxygen as reaction products. Sounds cool, but there is a whole range of technical challenges to overcome before we can even think of widespread commercialisation. The solar-to-chemical conversion efficiency is bad due to poor light utilisation, low density of the catalytically active sites, and rapid photoexcited electron-hole pair recombination.

Although the commercial use of photocatalysis is currently limited to a small number of processes, breakthroughs in reactors, as well as catalyst design, could potentially change that. For example, light-driven catalysts were capable of cracking ammonia into hydrogen at the same efficiency as thermal catalysts. This result could be promising for photocatalytic green ammonia production too. Another study found that a combination of reactor design and the use of CO2 as a scavenging agent significantly improved the NH3 formation rate. Even though these results were obtained at a lab-scale, they could be meaningful at a commercial scale too given the right financial as well as engineering support.

🚀 Potential advances to be made

  • Improving photocatalytic reactor design to improve light utilisation as well as electron-hole separation.
  • Improving photocatalyst materials by increasing the density of active sites and/or inhibiting electron-hole pair recombination.
  • Tweaking reaction conditions, such as temperature and pressure. These parameters could, for example, be used to suppress electron-hole recombination.

Biocatalysis

Nature has been fixating nitrogen for millions of years. Biological nitrogen fixation is done by a group of enzymes called metalloenzyme nitrogenases. These are composed of a bimetallic cofactor, such as iron-molybdenum (FeMo), as an active site and essentially convert nitrogen and protons into ammonia and hydrogen. As an energy source, the nitrogenases consume adenosine triphosphate (ATP), which is the same molecule that fuels our muscles. Ammonia doesn’t come for free though: for every ammonia molecule produced, the enzymes require 8 ATP molecules. In practice, nitrogen from the air diffuses into soil and bacteria (containing the nitrogenase enzymes) in the soil then convert it to plant-available organic forms. Agriculture leverages this process too: some plants have symbiotic relationships with nitrogen-fixing bacteria. Legume roots, for example, have nodules in which these bacteria can operate.

Instead of occurring naturally at the plants’ roots, this process can be done in a bioreactor as well. Certain (genetically modified) bacteria or enzymes could synthesise ammonia directly from water and nitrogen in the air. As such, they would mimic natural biological nitrogen fixation.

What does it cost?

Sounds like an elegant solution, but the downside is that even in bioreactor settings, the energy requirements remain high: about an order of magnitude higher compared to Haber-Bosch, which will result in much higher costs. Other challenges are that the microorganisms are susceptible to deactivation by oxygen and biological ammonia production is generally a slow process. Therefore, the biological routes alone simply cannot satisfy our current and future needs.

Greenhouse gas emissions

Similar to direct electrolysis, the greenhouse gas emissions associated with green ammonia production through biological pathways will likely depend on the carbon intensity of the electricity used.

🚀 Potential advances to be made

Market map for green ammonia production. Credits: Extantia.

Auxiliary technologies

Air separation and purification

Typically, the Haber-Bosch process requires pure nitrogen (N2) and pure hydrogen (H2) as input gas streams. This may sound straightforward and simple but actually requires additional process steps. Pure N2 can be obtained by separating it from the air (O2/N2=20/80). Pure H2 is currently mostly obtained through steam methane reforming. Consequently, the output gas stream additionally contains CO2 as well as traces of carbon monoxide (CO) and typically requires purification before usage. However, obtaining pure H2 becomes less of a problem when we use water electrolysis to produce (green) hydrogen.

Assuming the pure H2 issue will be solved through the deployment of water electrolysis, another question arises: why don’t we use ambient air directly as an input stream to create ammonia? The main hurdle is that oxygen poisons the iron-based catalyst required in the Haber-Bosch process. In other words: if ambient air is used, the catalyst cannot produce ammonia (as effectively) anymore.

This means that air separation will be essential at least for most green ammonia technologies based on Haber-Bosch. The three most common methods to do this are cryogenic systems, membranes, and pressure swing adsorption (PSA). Cryogenic systems basically cool air until N2 becomes liquid and then extract it for storage, transportation, or direct usage. These operations are typically performed in a centralised and large-scale fashion. For small to medium-scale on-site generation of N2, membrane systems are more suitable. They are built to separate compressed air through hollow-fibre membranes. They work by forcing compressed air into a vessel which selectively releases oxygen, water vapour, and other impurities out of its side walls. The N2 flows through the centres of the hollow fibres and emerges as gas. The membrane systems have relatively low operating costs and yield an N2 purity of 95–99.5%. Pressure swing adsorption is typically the best choice for high-purity N2 (up to 99.9995%). This process uses adsorbents that specifically target the impurities in ambient air and subsequently yield very pure N2.

Turbines and engines that run on ammonia

Combustion engines that can run on ammonia have been around since the 1800s and were briefly popular during World War II when oil shortages were a problem. However, fossil fuels proved both cheaper and easier to work with. Another challenge is that ammonia burns slower and is harder to ignite than fossil fuels; most engines would need some diesel or hydrogen to be kickstarted. Moreover, leakage of unburnt ammonia can be toxic and such engines tend to produce e.g. N2O, which is a more potent greenhouse gas than CO2.

Major engine manufacturers, including German MAN Energy, Finnish Wartsila, and Swiss WinGD, are now developing ammonia-fueled engines and kits to retrofit old engines so they can run on ammonia, with the first commercial products expected to be on ships in 2024. Meanwhile, startups are also getting into the game. In the US, Aza Power Systems was launched to commercialise its own ammonia-powered engine technology. Power production companies, too, are developing turbines that run on ammonia for electricity production.

Exhaust catalysts

Acid rain, a phenomenon that was prominent a few decades ago, posed a significant environmental challenge. The primary culprits behind acid rain were exhaust gases, particularly nitrogen oxides (NOx), emitted by various sources such as combustion engines. However, a groundbreaking solution was discovered in the form of exhaust catalysts, which played a crucial role in mitigating this issue. By implementing these catalysts, harmful gases could be converted into harmless nitrogen gas (N2), effectively addressing the problem of acid rain.

Exhaust catalysts will play an essential role in enabling ammonia utilisation. One of the key concerns associated with engines that run on ammonia is the production of nitrogen oxides (NOx) and nitrous oxide emissions (N2O), both of which contribute to air pollution and climate change. However, advancements in exhaust catalyst technology have provided effective solutions to tackle these issues.

Similar to the catalytic converters used in combustion engines for hydrocarbon fuels, catalytic converters can be specifically designed for ammonia engines. These converters utilise a series of catalysts that enable various chemical reactions. Two key components in this system are the ammonia slip catalyst and the selective catalytic reduction (SCR) catalyst:

  1. Ammonia slip refers to the unreacted or residual ammonia. The ammonia slip catalyst is specifically designed to capture and convert this excess ammonia into harmless nitrogen gas (N2) and water (H2O) before it is released into the environment. By incorporating an ammonia slip catalyst into the exhaust system, the emission of unreacted ammonia can be minimised, ensuring compliance with emissions regulations and reducing the potential environmental impact.
  2. The SCR catalyst facilitates the selective reduction of nitrogen oxides (NOx), including nitrogen monoxide (NO) and nitrogen dioxide (NO2), into harmless nitrogen gas (N2) and water (H2O) through a chemical reaction.

Fuel cells that run on ammonia

Fuel cells can generate electricity from ammonia by combining it with air and an oxidising agent. The only by-products are water and nitrogen gas. Ammonia fuel cells have potential in a variety of applications, such as transportation, power generation, and portable devices. Ongoing research and development efforts are focused on improving the efficiency and affordability of ammonia fuel cells to enable their widespread adoption.

An example of a promising system is solid oxide fuel cells (SOFCs). The high temperatures (500–1000°C) enable efficient electrochemical reactions and allow the use of a solid ceramic electrolyte, such as zirconia (ZrO2), which exhibits good ionic conductivity at elevated temperatures Zirconia-based electrolytes, such as yttria-stabilised zirconia (YSZ), are commonly used in solid oxide fuel cells. YSZ exhibits high oxygen ion conductivity at high temperatures, making it suitable for efficient ion transport within the fuel cell.

However, there are certainly challenges to be solved, such as the high-temperature and alkaline nature of ammonia fuel cells, which pose challenges related to materials compatibility and stability. The development of suitable materials for electrodes, interconnects, and seals that can withstand these conditions is essential.

Another key challenge is the efficient and controlled decomposition of ammonia within the fuel cell to release hydrogen gas for electrochemical reactions. This process requires catalysts and careful management to prevent catalyst degradation and unwanted reactions.

Ammonia fuel cells are an active area of research, and ongoing efforts are focused on improving their performance, durability, and reducing costs. Promising results have, for example, been achieved with specific nanoparticle coatings on YSZ anodes. These fuel cells hold promise for various applications, including stationary power generation, portable power systems, and potential integration with ammonia production plants.

Ammonia cracking (to release the hydrogen)

When ammonia is used as a hydrogen carrier, the hydrogen needs to be liberated from the carrier before usage. This can be done through a catalytic process. Via this process, the ammonia molecule is broken down, enabling the release of hydrogen gas. Catalytic ammonia cracking offers several advantages, including high conversion efficiency and selectivity. By employing suitable catalysts, the process can be optimised to achieve efficient and controlled hydrogen delivery. Ammonia cracking technologies are being actively researched and developed to unlock the full potential of ammonia as a clean and sustainable hydrogen carrier.

Improved efficiency of ammonia usage

About 80% of Haber-Bosch ammonia is used as agricultural fertiliser. In 2005, approximately 100 Mt NH3 from the Haber–Bosch process was used in global agriculture, whereas only 17 Mt NH3 were eventually consumed by humans in crop, dairy, and meat products. This highlights an extremely low nitrogen-use efficiency in agriculture (the amount of nitrogen retrieved in food produced per unit of nitrogen applied). About 40% of the ammonia fertiliser is converted back into nitrogen through microbial processes. Although harmless to the environment, it represents a waste of energy that was used for the Haber-Bosch process. The remaining unused fertiliser can be released as volatile NH3, can leach into natural reservoirs in the form of nitrates (NO3-), and/or eventually denitrify and form N2O, a potent greenhouse gas.

Improved agricultural management could be forged through increasing the efficiency of nitrogen usage in food production as well as improving the collection and repurposing of animal and human waste.

For example, increasing the nitrogen usage efficiency could be done through controlled/slow release of fertiliser. An example of this is the microencapsulation of the fertiliser for a controlled release process. What happens when ammonia is put into the soil is mostly microbiology: ammonia is converted into nitrite (NO2-) and eventually nitrate (NO3-) by microorganisms. Nitrate is the molecule that is the most biologically available to plants. The two main contributors to the so-called nitrification process are:

  • Ammonia-oxidising bacteria (AOB): these require high ammonia concentrations and grow at a relatively fast rate. Moreover, they produce significant amounts of N2O during growth.
  • Ammonia-oxidising archea (AOA): these require low ammonia concentrations and grow at a relatively slow rate. Their N2O production is relatively low.
  • By keeping the ammonia amounts and concentrations in the soil relatively low, the AOA contribution can be increased and plant growth can be supported without nitrate leaching while minimising the N2O production. An interesting side note to AOA is that they are very good at fixing CO2: they are 25% more efficient at it than green plants.

Chemical looping and waste valorisation

Chemical looping aims to take by-products from industrial processes and repurpose them. This could directly mean ammonia, but also hydrogen. Additionally, ammonia could be recovered from water or animal manure. Although this is an interesting pathway, the quantity obtained through this is (likely) not high enough to satisfy global needs completely.

Other considerations and logistics

  • Safety concerns around handling potentially toxic ammonia fuels.
  • Storage and transportation.
  • Water desalination may be required when green hydrogen is produced via water electrolysis.
  • Plasma-assisted ammonia production could decrease the energy input.
Market map outlining companies enabling green ammonia usage. Credits: Extantia.

The current state of implementation

Although that sounds promising, only tiny amounts of green ammonia are actually being produced at the moment. For example, a trial plant at the Fukushima Renewable Energy Institute in Japan uses solar power and water electrolysis to produce 20–50 kg of green ammonia per day. Another example is the demonstration system at the Rutherford Appleton Laboratory, in Oxfordshire, England, which is powered by an on-site wind turbine and makes up to 30 kg of green ammonia daily. These projects serve more as proof-of-concept and will help refine the technology before scaling up to larger production facilities.

According to the international energy agency (IEA), existing and announced projects for near-zero-emission ammonia make up a total of 8 Mt by 2030, which would only be 3% of the total (non-green) capacity in 2020. While ammonia production currently emits around 500 Mt CO2e annually, the Net Zero in 2050 scenario would only allow 17 Mt CO2e to be emitted by ammonia production annually. All in all: there’s a lot of work to do.

For example, research and development to advance green ammonia technologies are certainly still ongoing. This includes improving electrolysis efficiency, exploring alternative production pathways, optimising the utilisation of renewable energy, and investigating carbon capture and utilisation methods to further reduce emissions in the production process. International collaboration is crucial to driving the implementation of green ammonia technologies. Countries like Japan, Australia, Germany, and the Netherlands are actively engaged in partnerships and initiatives to promote the production, distribution, and utilisation of green ammonia. These collaborations facilitate knowledge sharing, technology transfer, and the development of common standards and regulations.

While green ammonia technologies are still in the early stages of implementation, there are several ongoing projects, research efforts, and policy initiatives aimed at advancing this field. The current production scale is relatively small, but there is a growing recognition of the need for emissions reductions in the ammonia industry, and concerted efforts are underway to scale up green ammonia production and achieve the necessary emission reduction targets.

--

--

Iris ten Have
Extantia Capital

Head of Science at Extantia Capital || Chemist by training || Climate tech unicorn hunter by passion