"Offshore CO₂ Injection: Why the Compression Route Went from Three Stages to Four"

On China's first offshore carbon capture, utilisation and storage project, in the Enping 15-1 oilfield of the Pearl River Mouth basin, the project team states it took the compression route from three stages to four — a four-stage pressurisation with dual-path stabilisation scheme — to hold the CO₂ in the supercritical state at an injection pressure above 180 kg, with core equipment and instrumentation independently developed and domestic supply at 100% (2). The injection equipment is two CO₂ treatment and injection trains each stated at 240,000 standard cubic metres per day, using the first domestic supercritical CO₂ injection compressor, placed in operation on the platform in 2023 (3). Cumulative injection passed 300,000 tonnes by early 2026, against a stated plan of more than 1 million tonnes over ten years with about 200,000 tonnes of additional crude production (2).

This article sets out why an offshore injection route is staged differently from an onshore one, what the extra stage buys, what has to happen to the gas before the compressor sees it, what is published, and where the evidence stops.

Why does an offshore route need a different compression arrangement?

Because the marine environment changes the state of the gas between the compressor and the well, and the published reason is condensation.

The project team's own account states the difficulty directly: offshore CCUS is harder to implement than onshore, and in the southern sea area the temperature and humidity swings are large enough that the CO₂ is prone to condensation, which can cause system blockage and equipment corrosion (2).

That is a different failure mode from the one that drives an onshore injection route. On land, the design argument is about reaching the pressure at which the CO₂ is dense enough to be injected as a dense phase, and the equipment is sheltered. At sea, the equipment is on a platform deck, the ambient swings between a humid tropical day and a cool night, and the fluid sits close enough to its phase boundary that a change in ambient conditions moves it.

The response in the published account has two parts: one more compression stage, and a control arrangement to keep the machine stable through transients. The team also states that a purge procedure was added for start-up and that a dual-circuit system was built (2).

What does the extra stage buy?

It buys margin — the ability to keep the fluid in the supercritical state while the conditions around it move — and the published figure for that state is an injection pressure above 180 kg.

Taking a compression train from three stages to four does not raise the pressure a single stage can produce. It divides the same total pressure rise into more, smaller steps, which lowers the temperature rise per stage and lowers the inter-stage pressure ratio. On a duty where the fluid must not be allowed to cross its phase boundary inside the machine, more stages is the way to keep the compression path away from that boundary.

The published statement is that the four-stage arrangement holds the CO₂ supercritical at the injection pressure above 180 kg (2). The same account gives the process chain as capture–purification–pressurisation–injection (2), and a separate account of the project's commissioning gives the injection rate as 8 tonnes per hour into a reservoir at 1,200–1,600 m, through a dedicated injection well 3,243 m deep (4).

What has to be done to the gas before the compressor sees it?

It has to be dried and stripped, and the equipment that does it is treated as its own package rather than as compressor auxiliaries.

On the Enping 15-1 project, the commissioning sequence recorded for the injection equipment includes a 24-hour loaded running test, regeneration and activation of the molecular sieve, automatic switching tests on the molecular sieve, and system gas-intake loading commissioning before the injection system went into service (3). A molecular sieve bed is a dehydration and separation step in front of the compressor, and its regeneration cycle is a scheduled interruption that the compression route has to be designed around.

The second offshore project takes the same idea further and packages it. On the Dongfang 1-1 gas field project, the equipment described as the project's core production-treatment unit — the country's first offshore low-temperature decarbonisation system — is built from three modules: a molecular-sieve adsorption unit for dehydration and mercury removal, a membrane-plus-low-temperature-rectification decarbonisation unit, and a refrigeration unit. It carries 26 items of individual equipment at a total weight of 2,252 tonnes, and its land-side build and shipment are described as an independent breakthrough in China's offshore oil-and-gas low-temperature decarbonisation package-building capability (1).

Read together, the two projects describe the same architecture: a pretreatment package that conditions the gas, and a compression train that raises it to injection pressure. The difference between them is that the first was an injection system added to an existing platform, and the second is a capture-and-injection loop built as a platform of its own — the new platform is the first three-legged jacket platform the company has deployed in the Hainan sea area, with a 1,400-tonne jacket and 1,100-tonne topsides, installed as separate lifts (1).

What the published parameters say

The stage count is the number to read first. A four-stage train on a duty whose injection pressure is stated in the region of 18 MPa is a ratio spread across four steps, and the published reason for the fourth step is the need to hold the phase rather than to reach the pressure.

Who has put what on the record?

The first domestic supercritical CO₂ injection compressor. Supplied by an engineering company's special-equipment arm and self-designed, self-built and self-commissioned, it went into service on the Enping 15-1 platform, forming two CO₂ treatment and injection trains. The account describes it as closing a domestic gap in offshore CO₂ storage equipment and as forming a complete offshore CO₂ storage equipment set with on-site service capability (3).

The second offshore project's pretreatment package. The low-temperature decarbonisation system for Dongfang 1-1, with its three modules and 26 items of equipment at 2,252 tonnes, described as built on land and shipped (1).

The two projects' scope. Enping 15-1 is a high-CO₂ oilfield development, where the associated gas contains up to 95% CO₂ and the injection is described as a carbon-drive, oil-store model — CO₂ injected to raise crude recovery while being permanently stored (2)(4). Dongfang 1-1 is described as the country's first offshore demonstration of injecting carbon to boost gas recovery, with an in-place offshore loop of capture and purification, formation storage and enhanced gas recovery (1). They are different duties in different basins, sharing the architecture and not the specification.

Where does this route stop?

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

"Above 180 kg" is stated as a scheme target, not a measured wellhead value. It appears in the description of the four-stage arrangement, and no measured injection pressure, discharge temperature or wellhead pressure is published alongside it.

The 300,000 tonnes is cumulative injection, and the 1 million tonnes is a plan. They are the only two production figures in the record for the first project, and neither comes with an availability figure, a compression-station uptime, or a maintenance interval for the injection equipment.

100% domestic is a statement about a programme. It is given as domestic supply of core equipment and instrumentation, and as a 100% equipment localisation rate. No bill of materials, no supplier list and no factory test report is published, so the claim cannot be checked item by item from public sources.

The second project has not been commissioned. Dongfang 1-1 has installed its platform and shipped its pretreatment package. Its stated 1 million tonnes a year is a full-operation expectation, and no date is given for first injection or commercial operation. Its compression equipment is not described at all — no machine class, stage count, pressure or drive appears in the published account.

And the compression route's parameters are not published. Across both projects, the record gives stage counts, a rate, a capacity per train and a target pressure. It does not give the compressor model, the drive, the motor rating, the suction pressure, the inter-stage pressures or the discharge temperature. The fourth stage is the only design decision that has been explained in public.

What does the trend add up to?

Three movements are visible, and they are consistent.

The constraint moved from the reservoir to the deck. The stage count was changed for a reason that comes from the marine ambient, not from the formation, which is the point at which an offshore compression route stops being an onshore route in a different place.

The pretreatment moved from being a set of auxiliaries to being a package. A 2,252-tonne, 26-item decarbonisation system shipped as three modules is not a dryer installed next to a compressor; it is a unit operation with its own weight, its own footprint and its own construction programme.

And the equipment moved from imported to domestic in one step. The first project's injection compressor is described as a domestic first and as arriving with 100% domestic core equipment and instrumentation, which is the shortest path from nothing to an equipment set in this record.

What has not moved is the evidence type. The cumulative injection figure is a production number, and everything about the machines behind it is a design statement, a stage count or a localisation rate.

What this article does not cover

Note that this article does not select a compressor, size a stage count, calculate a phase boundary, a discharge temperature or a pressure ratio, specify a material, a seal, a dryer or a molecular sieve, or give prices and delivery times. It describes the offshore duty, the published reason for the stage change, the pretreatment scope and the boundary of that evidence. It is not a purchase specification and it carries no compliance determination.

Where the next constraint sits

The question this article leaves open is where the boundary of a dense-phase CO₂ route actually sits — the phase behaviour itself, and how the target state sets the pressure rather than the reverse. That is worked in Compressing CO₂: The Pressure Is Set by the Phase, Not by the Distance, which takes the same fluid onshore and asks what changes between an oilfield flood, a saline aquifer and a pipeline.

The other end of the same route is the machine. Supercritical CO₂ Compression: What an Eight-Stage Integrally Geared Machine Changes sets out what a purpose-built dense-phase compressor train has to solve in its aerodynamic design, its rotor and its impeller, which is the layer below the stage count discussed here.

Sources