We Connect Asia

+6013-404 3889
bonjour@weconnect.asia
Menara Prestige, KLCC, Kuala Lumpur
Linkedin-in
GREEN AMMONIA

Green Ammonia Production Process:
From Water and Air to Ammonia, Step by Step

7 min read  ·  Nov 3, 2025  ·  By Alex

Ammonia production process: along a heavy industrial pipe rack that carries the whole ammonia production sequence in one line of sight.

The green ammonia production process has five steps: treat water, separate nitrogen from air, split water into hydrogen, react hydrogen with nitrogen, then store and load the ammonia. In practice, each step has its own equipment, its own schedule and its own way of delaying a project. If you can read the ammonia production process, you can read a technical review. And you can see where a developer carries exposure before a lender does.

In short

  • The ammonia production process is a chain. A weak link at any step slows everything downstream.
  • Only the hydrogen step is new at scale. The synthesis step is old and proven.
  • The step where the process meets the power supply is the one that needs the most design attention.
  • Storage and loading connect the process to the buyer’s specification.
Facebook
Twitter
LinkedIn

In this article

The ammonia production process in one paragraph

The plant first purifies its water, because an electrolyser needs clean feed. Next, electricity splits that water into hydrogen and oxygen. Meanwhile, an air separation unit pulls nitrogen out of the air. Then a compressor brings both gases to synthesis pressure, and a catalyst turns them into ammonia. Finally, the process cools the ammonia to a liquid and stores it until a ship, truck or pipeline takes it.

So that is the whole route. Everything in a green plant is a variation on it. For the equipment view of the same route, I wrote about what sits inside a green ammonia plant. This article follows the flow instead.

Step 1 and 2 of the ammonia production process: water and nitrogen

Water comes first. The plant needs treated water at a steady rate, and a permit has to cover the source. In practice, this is the step developers underestimate. For example, a site near the coast may need desalination. By contrast, an inland site may need an abstraction licence. Either way, both take time.

Nitrogen is the easy half. An air separation unit is standard industrial equipment, and suppliers deliver it as a package. However, the unit needs electricity too, and the design matches it to the hydrogen flow. So it should be specified together with the synthesis loop, not bought on its own.

For a schedule, steps 1 and 2 matter because they run in parallel with the long-lead items. Still, water permits and the air separation unit rarely set the critical path. But they can delay commissioning if nobody owns them.

Step 3: electrolysis and the hydrogen buffer

In the ammonia production process, electrolysis is the step that makes the plant green. The electrolyser turns water and renewable power into hydrogen, and the hydrogen then feeds the rest of the process. Also, this is the newest and the most expensive step: the electrolyser’s capex and operation dominate the cost of green ammonia (Nayak-Luke et al.).

Because renewable power varies, many plants add a hydrogen buffer. Designers size the electrolyser above the average need and store surplus hydrogen. As a result, the next step can run steadily even when the power dips. At the best locations, the process flexibility this needs raises the levelised cost of ammonia by 56% (Nayak-Luke et al.).

Above all, this is where design choices turn into money. First, a bigger buffer costs capex and land. Second, a smaller one asks more from the synthesis loop. So I look at this step first when I read a model.

Step 4: synthesis, where the ammonia production process meets the grid

Synthesis is the Haber-Bosch step. Operators compress and heat hydrogen and nitrogen, pass them over a catalyst, and the ammonia condenses out. In fact, the chemistry is mature, and that is its main strength.

However, this is also the step where the ammonia production process meets the power supply. Today, conventional loops run best at steady load. But a green plant feeds them from a variable source. If the loop cannot turn down far enough, the plant has to waste power or store more hydrogen. Either way, it costs money.

So the questions for step 4 are simple. What turndown does the loop support? How fast can it ramp? What does the supplier guarantee in writing? A lender will ask all three. Moreover, the answers feed back into the buffer size in step 3, which feeds back into the capex.

Step 5: storage, loading and the buyer's spec

Finally, the last step turns the product into something a buyer can take. Operators cool or pressurise the ammonia and store it in tanks. Then it moves out by ship, truck or pipeline.

This step of the ammonia production process is not just logistics. It is where the process meets the contract. Here, the buyer’s specification sets purity, the delivery window and the volume. If the plant cannot meet it consistently, the offtake is at risk, and so is the financing.

In practice, for plants in Southeast Asia, the loading route deserves early attention. Also, a jetty or a port slot is a project in itself. I look at the market side of this in the Singapore bunkering article.

What each step of the ammonia production process costs in energy

The synthesis loop sets the energy bill. At full load it uses 2.2 GJ per tonne of ammonia, and at low load the same loop uses 14.4 GJ per tonne (IRENA and Ammonia Energy Association). Therefore ramping down is never free, and a model that assumes constant load hides the largest variable in the operating cost.

Water costs less than it looks. On a coastal site, desalination adds at most about USD 0.02 per kilogram of hydrogen produced (IRENA and Ammonia Energy Association). In practice the water risk is rarely the bill. It is the permit, the discharge consent and the month when somebody finally applies for them.

Regional demand gives the schedule a deadline. Indonesia’s national roadmap puts ammonia demand at 1.3 million tonnes in 2030 and 4.1 million tonnes in 2040 (Indonesia’s hydrogen and ammonia roadmap). Pupuk Indonesia attaches green steps to existing plants rather than building from nothing. Hybrid green ammonia at Aceh carries 33,000 tonnes a year for 2027, and Gresik adds 82,500 tonnes a year for 2028 (Indonesia’s hydrogen and ammonia roadmap). So a new entrant competes with retrofit capacity that already holds the site, the water permit and the jetty.

Where the ammonia production process slips in practice

In practice, the ammonia production process rarely fails on chemistry. Instead, it slips on interfaces. Here is where it usually slips:

  1. Water permit. The team assumes the source rather than securing it, and the permit date stays open.
  1. Electrolyser delivery. The supply contract has no firm date or performance guarantee.
  1. Loop flexibility. The turndown assumed in the model does not match what the supplier will guarantee.
  1. Power profile. The renewable supply behaves differently from the model, and the buffer is wrong.
  1. Loading. The plant reaches readiness while the jetty or pipeline still lacks funding and permits.

So because these steps depend on each other, a delay in one usually moves the others. For example, a late loop guarantee delays the lenders, which delays the equipment order.

In practice, the order of work matters as much as the work itself. Water and its permit come first, the power contract second, the electrolyser delivery date third, the loop guarantee in writing fourth, and the loading route fifth. The offtake runs alongside the power from day one, and teams that sequence the interfaces this way spend far less time re-running the model.

Checklist

  • The team has secured the water source and the permit.
  • Designers size the air separation unit together with the synthesis loop.
  • The electrolyser contract states delivery date and performance.
  • The model shows a buffer matched to the real power profile.
  • The loop supplier confirms turndown and ramp in writing.
  • Permits and funding cover the loading route.
  • The design meets the buyer’s specification, not only the contract.

If the team leaves most of these open, the next step is not another model run. It is closing the interfaces, starting with the offtake and the power supply.

Next step

If your project is moving toward FID and the offtake or the capital is the open point, book a 30-minute call or send me your project brief. I tell you early if the project is not ready. You can also look at how the ammonia route can start from an existing plant.

Sources

FAQ

Water treatment, nitrogen separation, electrolysis, ammonia synthesis, then storage and loading. Each step has its own equipment and schedule.

The hydrogen. It comes from water split with renewable power, not from natural gas.

Because the loop prefers steady load and renewable power varies. A hydrogen buffer or a flexible loop has to bridge the gap.

The electrolyser and the renewable supply usually set the critical path. Permits for water and loading can still delay commissioning.

At the interfaces between steps: power, electrolyser delivery, loop guarantees and loading. The chemistry is rarely the issue.

Lenders want each interface covered by a contract or a guarantee. A step with an open interface is a step they will price or refuse.

Not sure your project would survive a credit conversation?

A short call is usually enough to see where your project stands against what a lender or an equity investor will ask for. Tell me where your project is today, and I will tell you what the next conversation needs to look like.

Book a discovery call
Alex, Founder of WeConnect Asia

Alex, Founder of WeConnect Asia

I help low-carbon fuel and carbon removal project developers in Southeast Asia secure the commercial and financial partners needed to reach FID.

More about me·Get in touch

Facebook
Twitter
LinkedIn

Related articles