"Green Ammonia Synthesis Gas: What the Hydrogen in the Feed Changes"

A green ammonia plant has one machine whose duty is defined by the gas rather than by the product. At the Songyuan hydrogen industry park — a green hydrogen–ammonia–methanol integrated project in Jilin whose first phase started operation on 16 December 2025 with 800 MW of new-energy generation and a stated output of 45,000 tonnes of green hydrogen and 200,000 tonnes of green ammonia and methanol a year — the synthesis gas compressor units were independently developed and manufactured by the supplier, and the company describes them as the first machines of their class in China's synthetic-ammonia sector, with first-time commissioning success (1)(2). The design basis published for those units is 5 years of continuous operation and a unit design life of not less than 20 years (1).

This article sets out what that machine does, why the presence of hydrogen rather than the synthesis pressure is the technical point, what the load-following requirement adds, and where the published evidence stops.

What does the synthesis gas compressor do in a green ammonia plant?

It is the link between hydrogen production and ammonia–methanol synthesis, and its published task is to raise the feed gas to the pressure the synthesis reaction needs (1).

That places it downstream of the electrolysers and downstream of the gas holder or buffer. In the Songyuan project the process route is described as wind-and-solar direct supply, storage for peak shaving, a high-efficiency electrolysis system and dynamic ammonia synthesis integrated together, with the ammonia and methanol produced from green hydrogen (2).

The distinction that matters for the machine is what the feed gas is. It is not hydrogen — it is a hydrogen-bearing mixture heading for a reactor, and the design has to hold that composition steady enough for the catalyst while its supply rate moves.

Why does the hydrogen content, rather than the discharge pressure, set the machine?

Because a gas whose molecule is the lightest one in industrial service behaves unlike any other process gas at the same volume flow, and the published problem list names that behaviour first.

The problems the supplier states its engineering team solved are: efficient aerodynamic design for hydrogen-bearing synthesis gas; long-term operational reliability under extreme conditions; mechanical stability under high torque; and rotor dynamic stability for a large machine (1).

Four items, and three of them are consequences of the gas rather than of the pressure. Hydrogen's molecular weight is 2.02, so a given volumetric flow carries a small mass flow, which means a large machine for the mass delivered and a compression duty that is expensive per kilogram. The same small molecule leaks past clearances and seals more readily than heavier gases, so internal leakage and sealing become design items rather than maintenance items. And because the gas arrives from an electrolysis train rather than from a reformer, its supply rate is a function of the weather.

The published account puts the load-following requirement in the same list as the mechanical ones, which is the honest way to state it: this is not one problem with a gas and a separate problem with the power supply. It is one specification.

What does a variable power supply add?

It converts an inlet condition that would otherwise be a design point into a moving one, and it is the reason the project's own control system is part of the equipment story.

The Songyuan project states that it developed a large-scale integrated intelligent control system for wind, solar, storage, hydrogen, ammonia and methanol, which uses forecasts of wind and solar resource together with real-time power output to adjust the chemical plant's production load dynamically — described as load following the source. The stated purpose is that the random, fluctuating output of wind and solar generation would otherwise break the continuity of the hydrogen feed, which the project identifies as one of the main bottlenecks for large-scale application of green hydrogen to ammonia and methanol (3).

For the compressor this means the machine cannot be specified only for its rated point. The supplier states the units were designed to keep running in a scenario of unstable hydrogen supply caused by wind and solar fluctuation, and to follow load changes in continuous ammonia production (1).

The same project states four world records for the configuration — largest scale, largest hydrogen storage capacity, widest load-flexible process, and largest alkaline electrolysis equipment — and states that key core equipment reached 100% domestic supply, with more than a hundred patents, proprietary technologies and standards generated (1)(2).

The machine classes that meet on this route

A green ammonia plant does not use one compressor class, and the split is a useful way to read the published record.

Read down that table and a boundary appears. Where the volume flow is large and the pressure rise is delivered in one continuous machine, the published examples are centrifugal trains. Where the duty is a packaged or distributed unit, the published example is a diaphragm machine, and that machine is described by its industry association as holding the feed side of distributed green hydrogen, green ammonia and green methanol production (5).

That is a normal division rather than a competition. The diaphragm machine's advantage is containment — a static seal and a process gas that never reaches hydraulic fluid. The centrifugal train's advantage is flow at low ratio in a single casing. On the feed side of a large synthesis loop, flow is what is being bought.

Who has put what on the record?

Songyuan, Jilin — the first large-scale green hydrogen–ammonia–methanol co-production project. Its first phase is stated at 45,000 tonnes of green hydrogen and 200,000 tonnes of green ammonia and methanol a year on 800 MW of generation, in operation from 16 December 2025, with ISCC EU green certification and a stated first global sales contract for green ammonia as ocean-going fuel. The stated annual savings are about 600,000 tonnes of standard coal and 740,000 tonnes of CO₂ (1)(2).

Envision Energy's green ammonia plant. The world's largest zero-carbon green ammonia plant is given as 1.52 million tonnes a year, with syngas compressor trains supplied and the full plant commissioned (4).

The Jiangxi ammonia project. An 800,000 t/a ammonia synthesis project, for which the supplier states its in-house syngas compressor train passed first-time commissioning and outperformed imported equivalents on the key energy-efficiency indices (4).

The Australian green-ammonia shipment. Three large diaphragm compressors delivered and shipped for an Australian green nitrogen-fertiliser project, with the supplier stating a merged design across AS3000, IEC, ASME, AS1210 and the Australian steel structures code, passing international third-party inspection on the first submission (5).

The green methanol line. A 1.45-million-tonne-a-year green methanol project in Inner Mongolia, structured in three phases, with the 250,000 t/a first phase in production and its first product shipment completed on 10 September 2026 and sent onward to the port of Busan. The project is described as running wind-and-solar generation, electrolytic hydrogen, low-carbon biomass gasification and green-hydrogen-coupled methanol synthesis in one chain, with a stated green-electricity share above 75% (6).

Where does this route stop?

At five places, all of them visible in the published record.

The five-year and twenty-year figures are design requirements, not service records. They are stated as the design basis of the units. The Songyuan project began operation on 16 December 2025, so the longest possible operational record behind these machines is measured in months, and no measured availability, overhaul interval or mean time between failures is published for them.

"First machine of its class in China" is a statement about a programme. The claim rests on the unit being the first of its kind in the domestic synthetic-ammonia sector, and on a first-time commissioning. It is not a statement about the number of machines built, and no build count is published.

The energy-efficiency comparison is a company statement. The Jiangxi project's "better than imported equivalents on the key energy-efficiency indices" is the supplier's own comparison. No third-party test report, no efficiency figure and no comparator are published alongside it.

100% domestic applies to key core equipment. The project states key core equipment reached 100% domestic supply. That is a statement about the equipment scope so far identified as key, and it is not the same claim as a fully domestic bill of materials.

The load-following capability is stated as a design intent. The control system and the units are described as built to follow a fluctuating supply. No measured turndown range, ramp rate or number of start-stop cycles is published for the compressor side, so the boundary between "designed to follow load" and "demonstrated to follow load" is not closed by these sources.

What does the trend add up to?

Three movements are visible, and they are consistent.

The machine moved from being specified as a gas-handling device to being specified as part of a control loop. The published problem list for the Songyuan units contains a rotor-dynamics item and a load-following item in the same breath, which is what happens when the feed supply is a weather variable rather than a contract.

The duty moved upstream of the reactor rather than upstream of the customer. The green ammonia and green methanol projects in this record are all selling a chemical or a fuel, so the feed gas has to be held close enough to a catalyst specification that the reactor does not have to absorb the swings.

And the scale moved. A 1.45-million-tonne green methanol project and a 1.52-million-tonne green ammonia plant are not pilot units, and the machine classes that serve them are the same classes that serve conventional ammonia, which is a sign that the equipment question on this route has already been answered by the conventional industry.

What has not moved is the evidence type. Every figure in this article is a design value, a stated capacity or a supplier's comparison, and the operating windows behind them are months long.

What this article does not cover

Note that this article does not select a compressor, size a train, calculate a pressure ratio or a discharge temperature, specify a material, a seal, a diaphragm or a rotor, or give prices and delivery times. It describes the duty, the published problem list, the machine classes that appear in the record and the boundary of that evidence. It is not a purchase specification and it carries no compliance determination.

Where the next constraint sits

The feed-gas route in this article ends where the synthesis loop begins, and the constraint that follows it is the one the storage side has to carry: a green ammonia plant that follows the weather is only as good as its ability to hold hydrogen when the wind drops. Salt-Cavern Hydrogen Storage: Why the Store Sets the Compressor's Duty sets out what a geological hydrogen store asks of its compression equipment.

The other adjacent question is what happens when the same gas has to be compressed somewhere far less forgiving than a chemical plant site, which is the subject of Offshore CO₂ Injection: Why the Compression Route Went from Three Stages to Four — a machine duty set by the environment rather than by the gas alone.

Sources